A New Power System

[unmerged(6147)]

A SUGGESTION FOR A REVISED POWER SYSTEM

As most of you know there is a power point part on Tonio’s great Stats Page. It was introduced for the campaign Test of Skill.

After having used it for a few months several ideas for improvement have occurred to me. Some of the problems with the current system are

1. The estimation of potential/future power (as in manus who generate wealth) as opposed to current power (as in armies and ships) is somewhat arbitrary. When I made the original system I e.g. gave manus almost no value at all late in the game. I believe a better approach is to view the game as not ending in 1819, i.e. the power of certain features should not decrease just because the game is about to end.

2. The power value assessed to most aspects (ships, manus, armies, CCs etc) is also rather arbitrarily and subjective (my own view). It appears preferable if we could more objectively assess a similar value to all assets. That would be possible if we transformed as many assets into money. This is not made today.

3. Tonio is presently enlarging the scope of the power presentation. Instead of only presenting the value for the latest session he now also gives an average value for several sessions. However, in our current power system many features (like income) have its power value increased with time. For an average value this means that earlier sessions become much less important than later ones. Thus we need to modify more aspects of power depending on the current year. If you earn say 1000d in they year 1500 that should perhaps result in as much power points you get from earning 10,000d in 1819.


Below I will deal with each aspect of power. But first I will start with some common things and then describe each aspect of power for it itself, both presenting the current formula and the revised one.



COMMON THINGS


YEAR MODIFIER

Through out I will use a year modifier to solve problem 3 above.

When analysing games it can be found that income increase with about 0.8% per year. However the rate of expansion is not linear. In some periods – especially early on - income increase more rapidly than else. If we set a base value of 1 for the year 1500 (well, really 1499) realistic mulitpliers used (assuming the multiplier aims at transforming values from different eras into comparable values) will look something like this
< 1450 2
< 1475 1.5
< 1500 1
< 1525 0.8
< 1550 0.6
< 1575 0.5
< 1600 0.4
< 1625 0.35
< 1650 0.3
< 1675 0.28
< 1700 0.25
< 1725 0.23
< 1750 0.2
< 1775 0.19
< 1800 0.17
Later 0.15

Example: if a session ends say in 1499 then 1 is used as year multiplier, if it ends in 1776 0.19 is used as multiplier. Assume income in 1776 is 10,000d per year. Then it will be multiplied by 0.19 and thus give a value of 1,900. Thus we can make a fair comparison of income from two different years. Of course, this value must then be transformed into power points by more adjustments.

An alternative would be for Tonio’s program to count the number of years played since start and the compute the year modifier based upon an “inflation” of 0.8% per year. That would be give a more accurate value for years far from the middle of my intervals but would less accurately portray the more rapid expansion early on.

You will find this year modifier used for almost all aspects of power we calculate.

I will call this modifier YM when presenting the formulas.


VALUE MODIFIER

After having computed an economic value for an aspect (be it manus or army etc) we must then transform that into power points. Presently I am contemplating giving 1 power point for each 500d in value. Please note that income is different from other aspects. As an example: would you prefer to have 9,000d in yearly income and assets (CCs, shipyards, manus, armies etc) worth 1000d or would you prefer income of 1,000d and assets of 9,000? Well, of course you would prefer the former alternative. Thus the value modifier for income will be special, presently I am contemplating giving 1 power point for each 50d.

I will call this modifier VM when presenting the formulas.


INFLATION MODIFIER

In all applicable cases we will multiply the value of the aspect with a decimal number portraying the inflation of the nation in question. Assume inflation is 10%, then we will divide by 1.1.

I will call this modifier IM.


DETAILED ANALYSIS OF OUR DIFFERENT ASPECTS OF POWER


PP = power point.


A) INCOME

Previously it was:

For each 300d (adjusted by inflation) in yearly income (peace and event income not included) you get 1 PP.
Formula: ((All income – income from peace and events) / (1+ stored value of inflation)) / 300.

Now I propose:

For each 50d in yearly income (peace and event income not included) you get 1 PP after adjustments that are IM, YM and the special AM for income.
Formula: ((All income – income from peace and events) / IM) * YM /VM

Example: Year is 1819. Yearly income except event and peace is 10,000. Inflation is 10%.
10,000/1.1 * 0.15 / 50 = 27,2


B) CURRENT MP POOL

Note that the cap of the MP pool is unfortunately not stored in the save and instead we have to do with the current value. This is bad but the size of the MP pool is such an important thing that we cannot just disregard it.

Previously it was:
For each 25,000 men in your current MP pool you get 1 PP.
Formula: Current MP pool value / 25000

Now I propose
You get 1 PP for each 4,000 in your MP pool after applying the year modifier.
Formula: Current MP pool * YM / 4000
Example: year is 1819 and the MP pool is 700,000.
700,000*0.15/4000 = 26.25

As you can see we in this case use the year modifier for a non-economic value. This is not ideal as the speed in increasing MP does not exactly mimic the eco growth but it is not that far away and for simplicity it has been chosen. Incidentally this is a big change compared to how it is today in that the value for the MP pool will be much higher for early sessions. However for late sessions the value will be just about what it was before.


C) ARMY AND NAVY MORALE

Previously it was:

For each 0.1 higher morale at 100% maintenance your army has compared to the one human player who has the lowest such morale you get 1 PP.
Same for navy morale.

Formula: (Your morale at 100% maintenance - the lowest human ditto) * 10.

As the values for many other aspects will in early sessions increase a lot with this new system I suggest a change. Also army morale is arguable more important than naval morale. Thus I propose
a) you get 1 PP for every 0.05 higher army morale
b) you get 1 PP for every 0.1 higher naval morale


D) TECHS

Previously it was:

For each 10 "tech points" you get 1 PP.
For each infra/trade tech level you get 6 tech points. For each LT/NT you get 1 tech point.
Formula: (trade tech level * 10 + infra level * 10 + land tech level + naval tech level) / 10.

Note there is an error in this old formula, the eco techs should be multiplied with 6 not with 10, as can be understood when reading the text).

Now I propose
Same basic formula, but then applying the YM and then an appropriate weight factor (comparable to the division by 10 in the old formula. I am presently contemplating a weight factor of 0.5.
Formula: (trade level * 6 + infra level * 6 + land tech level + naval tech level) * YM * 0.5

Example: year is 1819 and you have maxed all techs.
(60+60+60+60) * 0.15 * 0.5 = 18

Note: for early sessions this will make tech much more important than before.


E) FLEET SIZE

Previously it was:
Formula:
Min(40,
(
( number of warships * (sum of your shock and fire value for war ships)
+ number of galleys * (sum of your shock and fire value for galleys)
)
/
sum of human tech leader's shock and fire values for war ships and galleys
/ 20
)

Now I propose:
((Number of galleys * cost for a galley for that nation) +
(Number of warships * cost for a war ship for that nation))
/IM *YM /VM
With a cap on 20.

Example: year is 1819 and you have 700 warships and no galleys, inflation at 10%. Each warship costs 60d.
(0 + 700*60) / 1.1. * 0.15 / 500 = 11.45


F) FORTRESSES

Previously it was: For each 1/10 of average fort level you have more than the one human player with the lowest average fort level you get 1 PP.
Formula: (Your average fort level - the lowest average fort level) * 10.

To build a fortress at offensive 5 and 0% inflation costs
level 1 = 125d
level 2 = 250d
level 3 = 375d
level 4 = 625d
level 5 = 1250d
level 6 = 2500d

As can be seen the ration of increase of cost for forts is not linear.

The true value of a certain fortress level must include the cost of building all minor levels as well. Thus the ratio between fortress investment in a provinces is:
level 1 = 1
level 2 = 3
level 3 = 6
level 4 = 11
level 5 = 21
level 6 = 41

Thus what we need to do when computing the dividend is not using the fortress level but their cost ratio. So if you have two provinces, one with a minimal fortress and one with a maximal fortress your average "fortress level" (i.e. fortress level when computing power points) should not be 3.5 but should be 21.

Thus I propose
( number of minimal fortresses * 1
+ number of small fortresses * 3
+ number of medium fortresses * 6
+ number of large fortresses * 11
+ number of mighty fortresses * 21
+ number of maximal fortresses * 41)
/ number of city provinces

Then we must decide upon how much PP you should get for having a better average of fortress level than the one with the lowest. We cannot any longer give 1 PP for every 0.1 higher since the differences will be much larger with this new system. There for I suggest: For each 0.25/10 of average fort level you have more than the one human player with the lowest average fort level you get 1 PP.

Formula: (Your average fort level - the lowest average fort level) * 4.


G) ARMY SIZE

Same algorithm as for fleets. But now we have three types, infantry, cavalry and guns.


H) SHIPYARDS

Previously it was: Formula: number of shipyards / 2

Shipyards presents a new problem. The cost for a shipyard is not static. The first costs 1000d and for each new one you have to pay 100d more than for the previous one. Thus the value for the shipyards is not that easy to calculate. If Tonio can compute a correct sum for having for example 6 shipyards (which is 1000+1100+1200+1300+1400+1500=7500d) that would be swell. Else I suggest using an average value of 1200 since the number of shipyards at the end perhaps usually is around 5. So if you have 5 shipyards they add up to 6000d using the average method formula.

I propose: value of shipyards / IM * YM / VM

Example: year is 1819 and you have 10% inflation and you have 5 shipyards and we use the average value of 1200.
5*1200 / 1.1 * 0.15 / 500 = 1.63


I) MANUS

Previously it was quite complicated.
Formula: (number of refineries + number of weapon * 0.5 + sum of other manus * 0.33) * year multiplier
Each refinery gives 1 manu point.
Each weapon manufactory gives 0.5 manu poinst
Other kind of manu give 0.33 points.
The year multiplier was a special one used to really get the late game value of manus down to almost nothing.

I you want to use the cost of building the manus, like we did for shipyards, instead of these arbitrary values previously used then we face an even more complicated problem than for shipyards because the basic cost algorithm for manus has because of its complexity never been officially analysed and published. But if we allow us to be a little simplistic we can say that basically each new manu costs 200d more than the previous one. For simplicity we will also stop make a difference between different kind of manus. So here again: if Tonio can compute an exact value we will use that, else we will use an average value. Assuming the number of manus you have at the end will be around 15-20 the average cost will be around 2000d a piece.

Thus I propose:

Formula: Manu value / IM * YM / VM.
Example: Year is 1819, inflation at 10% and you have 16 manus. Average value of 2000d assumed.
16 * 2000 / 1.1. * 0.15 / 500 = 8.72


J) CCs

Previously it was: Formula: number of CC / 10

For CCs the initial cost is 1000d and then they increase with 250d per new CC. Again we have two alternatives: if Tonio can manage to compute the exact cost we use that, else we use an average value. Assuming you have around 15 CCs at the end the average cost will be around 2500.

Thus I propose:

Formula: CC value / IM * YM / VM.
Example: Year is 1819, inflation at 10% and you have 15 CCs. Average value of 2500d assumed.
15*2500 / 1.1 * 0.15 / 500 = 10.22


K) TREASURY

Previously it was: Treasury size / (1+ stored value of inflation) * year multiplier / 200
The year modifier used was the same as I propose now, although with values only for every 50th year instead of every 25th year.

Now I propose (as you will by now guess if you have read everything so far)
Treasury size / M * YM / VM.
Example: year is 1819, inflation at 10%, 5000d in treasury.
5000/1.1*0.5/500=4.54


L) CoT:s OWNED

Previously this was not implemented. I thought it was aptly portrayed by the income for the owner and, in case of embargo, of income from the embargoed. However they represent a big potential value that can be used at any time, i.e. even if you do not presently embargo a nation you may do so at any time (providing you do not have a truce or TA) and thus it should be taken into the power formula after all.

Ideally you should use the value of the COTs to calculate the power, however that is not accessible in the save, so we have to resort to simply counting them.

I propose you get 2 power points for each COT owned.

=====================

As can be seen this new proposed system is far from as subjective as the previous one was. All aspects that can be transformed to ducats are transformed alike with the exception for income that is valued higher than other aspects. This means that the system does not say it is smarter to buy refineries than FAAs, nor FAAs than shipyards etc. It simply calculates the worth. I believe this means a much better system than the previous one in which I tried to guess at a “true” power value for different aspects of power – sometimes ending up at arguably crazy values as Tonio posted about when I introduced the power value for treasury.

And lastly, it is possible to create a new category, ownership of CCs. That would be very simple I think, just give some power points for each CC, I do not think we should use any year modifier although you normally own fewer at the start. Because after some 150-200 of years the number of CCs owned by humans in the game almost stop to increase, they are almost all already taken.

-----------

If all players in ToS agree to this new system and Tonio have time and interest to implement it on his stats page it is my wish that we adopt it in ToS.

 

[Tonioz]

Daniel, i agree that points increasing with time increase slightly the difference between nations. But mostly it is allright, and i don`t think addional complicated modifiers required. But it is up to you, it is your rating system.

What do you think about grouping your rating to several groups? Like land power, naval power etc ?

 

[unmerged(6147)]

Yes grouping would be interesting. Although at the moment I think I will focus on the new changes I suggest.

You could perhaps split up in 3 groups

1. Land power
2. Naval power
3. Eco power

 

[King John]

Those groupings would be good.

I agree with Tonio on using year modifiers, what is the point? It shouldn't affect the relative rankings whether we use them or not, but it does do a bad thing by making it harder to measure your own progress against your previous scores. And for the sake of simplicity, it also makes sense not to use them.

The one thing that I think makes sense to use year modifiers with is MP, but in the exact opposite way you have it Daniel. Most of your manpower is useless at the beginning of the game, because you can't use it due to lack of income. So if you've got 100 MP, you shouldn't get a high PP score in 1450 for that, you should have the biggest negative modifier at this time. MP's value doesn't increase until you can use your MP. By about 1650, all modifiers on MP should be removed, and it should stay that way to the end of the game.




I think it'd be cool to also include a section on "potential growth", which could include the years left in the game, the quality of your coming monarchs, the tech levels left to achieve, the price of new manus, and the number of slots open to you in cots(up to 5). Your inflation level should play a part in this as well, as a good indicator of the amount of capital you can afford to raise for economic projects. Treasury and income should of course factor into all this as well.

 

[unmerged(6147)]

I agree with Tonio on using year modifiers, what is the point?

The point have been explicitly stated in the post I made: It makes it possible to more correctly compare power ratings for different years, which is very useful when using Tonio's average rating. As it means all sessions become more equally important. I envisage a future where the average method will be used as an objective method to crown the winner in MP games and it could be quite interesting for us in ToS to have that as an alternative way to judge the efforts. Of course, we did not know we had that possibility from start and thus it does not necessarily give a good picture of our efforts, but anyhow it can be fun for us. And those doing less well in the power rating after the end session can perhaps get some comfort from instead pointing at their end position on the average ranking.

Further more we already use year modifiers, one for treasury and one for manus. And no one complained about them.

It shouldn't affect the relative rankings whether we use them or not, but it does do a bad thing by making it harder to measure your own progress against your previous scores.

I do not understand why it should? I could understand someone arguing it would make it easier. Let us take an example. Say you have 40 PP after the previous session. Now say we use the old system and you end up at 42. Have your power increased. Well yes, it is has, but how much has it increased compared to what could be expected from you? Of that we know nothing. Now say we use the new system and your PP ends up at 40, the same value. Now we can see that your power has not increased compared to what would be expected. Which I as a player would be much more interested in knowing.

It can be compared to estimating your own economy. Say you have $100 on the bank on Jan 1 the year 2006. Then on Jan 1 2007 you have $101. Now you may be very happy and celebrate because your wealth has increased with 1%. Meanwhile your good old friend Daniel makes a similar analysis. He also had $100 on the bank on Jan 1 2006 and also have $101 on Jan 1 2007. However he reads the newspapers and such and thus knows that inflation during 2006 was exactly 1%. So Daniel tells himself: hey, my wealth did not increase at all, I just kept in step with inflation. Thus he saves his bottle of champagne for another year.

So, in fact, I believe my new power system makes it easier to estimate the progress of your nation, not more difficult as you claim. The new system not only makes it possible for you to judge your general progress (if you stay at the same power value your progress is about 0.8% per year, the normal value) but it also makes it possible to better compare your relative posittion, realtive to normal EU games that is. Your position relative your opponents in the game can be easily deducted by checking the Standings' post I make after each session.

And for the sake of simplicity, it also makes sense not to use them.

If you like simplicity you should applaud my suggestions for how to estimate fleet and army size.

The one thing that I think makes sense to use year modifiers with is MP, but in the exact opposite way you have it Daniel. Most of your manpower is useless at the beginning of the game, because you can't use it due to lack of income. So if you've got 100 MP, you shouldn't get a high PP score in 1450 for that, you should have the biggest negative modifier at this time. MP's value doesn't increase until you can use your MP. By about 1650, all modifiers on MP should be removed, and it should stay that way to the end of the game.

An intruiging idea. However quite complex. For a small nation early MP can be a problem. And for nations that get much financial help from other (think AUS getting subsidies from SPA) it definitely can. If one could construct a formula that better portrayed the real value of MP it would be very appetising.

I think it'd be cool to also include a section on "potential growth", which could include the years left in the game, the quality of your coming monarchs, the tech levels left to achieve, the price of new manus, and the number of slots open to you in cots(up to 5). Your inflation level should play a part in this as well, as a good indicator of the amount of capital you can afford to raise for economic projects. Treasury and income should of course factor into all this as well.

Well, you are imaginative today John! But these suggestions appear much too complex. Although very interesting indeed they have nothing do with "power" as I define it, i.e. the actual power just now. Indeed, I have just eliminated one of the few "potential power" parameters in the power formulas - that for the manus. Potential growth is so complicated. Take the world in 1489. Now SPA has a an enormous potential growth thanks to Columbus.... Now, how could Tonio's program be changed to accurately portray that without you and me going broke because of the wages we had to pay Tonio for his redesign of the program.

 

[unmerged(6147)]

Decreased value if income and MP slightly. Increased value of army morale.

 

[King John]

Daniel, will your system be accurate for all scenarios? 1419, 1453, 1520, unhistorical/historical? Unless its designed such that you can plug in the modifiers you want for the different scenarios yourself, its going to have irregularies.

But if the main point is to get average power ratings over the course of several sessions or the whole game, you could accomplish the same thing by weighing normal PPs each session: Add all PP scores together, then divide each individual score by the total to get percentages, then take each percentage multiplied by any number, maybe 250. Using these final numbers for however many sessions you want to find the average power ratings for every country, the end results should still add up to 250(unless some countries get eliminated in the interval). Thus you could get your averages, and this would more accurately show whether you lost or gained position than your idea. And you wouldn't need to screw up the normal power points.




As for MP, you have a good point about smaller countries. A better way to handle this would be to determine how much MP can be useful based on your yearly income and the price of troops. Perhaps something like this:

Basing the average troop ratio on 7/10 infantry, and 3/10 cavalry

Yearly income / (inf costs * 7 + cav costs * 3) * 10 = useful MP.

So with 1000D yearly income, 10D inf costs, 20D cav costs, you'd have

1000D/ (10 * 7 + 20 * 3) * 10 = 62.5 . So you would plug in up to 62.5 MP for PPs.

If a country can't afford more than that, it doesn't deserve much credit for the MP it can't support. Though you could give something at least in case it gets financial support. Perhaps multiply whatever spare MP the country has by .5, or less.

 

[unmerged(6147)]

Daniel, will your system be accurate for all scenarios? 1419, 1453, 1520, unhistorical/historical? Unless its designed such that you can plug in the modifiers you want for the different scenarios yourself, its going to have irregularies.

No, as I have said the increase in wealth is not linear. You increase more rapidly in the beginning than at the end, probably due to a series of effects, such as
- human wars are much more common in the end phase
- you gain fewer eco tech levels
- some stop colonising

But the correlation it is pretty good I say. You can look at the values we use today as the year modifier for treasury size, they are more accurate but have a big drawback, and that is that they are in intervals of 50 years, thus there can be big threshold effects when you end up a year on either side of the border between two intervals. I wanted to get rid of that. As a whole I hope the new year modifier I propose will work better.

BTW, Tonio has agreed to simply calcualte the number of years we have played from start and then assume an increase of wealth of 0.7% per year which appears to be a more accurate estimation than the former number of 0.8% I gave. I have today studied some 4 different campaigns. The best effort I spotted was Tonio's Prussia in C&C1 which on average had an increase of 1% per year. Then I also found some really terrible ones, like a FRA who only doubled her income once during some 310 years of play. Good players normally seem to double their income every 90-100 years or so. 0.7% per year equals 100 years and 0.8% equals 87 years.

If you make a scenario of your own, like we did, and generally increase/decrease tax values and production values, which I did not, etc you will of course get a worsened correlation between different campaings. In our campaign I looked through the province file before start and gave us values that I believed to be average values for Europe. Only the value of the gold mines were decreased but on the other hand we got a few COTs extra in Europe. All in all I think our values are pretty standard. For Watk3.1 that is.

But if the main point is to get average power ratings over the course of several sessions or the whole game, you could accomplish the same thing by weighing normal PPs each session: Add all PP scores together, then divide each individual score by the total to get percentages, then take each percentage multiplied by any number, maybe 250. Using these final numbers for however many sessions you want to find the average power ratings for every country, the end results should still add up to 250(unless some countries get eliminated in the interval). Thus you could get your averages, and this would more accurately show whether you lost or gained position than your idea.

Yes, one could do so. However,
1. it makes it more difficult to compare between games since your effort will be compared with the effort of your opponents, not with a more objective measurement, which is what I would like
2. I believe players prefer points rather than having them transformed into percentages, the more you transform the more far from reality you come and the more difficult for players to remember how they were made up, but I may be wrong in this
3. Much work for Tonio

If Tonio would agree to do this and themajoirty in our game would prefer it I have no serious objection.

Actually I fail to see what you have against this idea of a more accurate year modifier than the one we use for treasury size today and to use it for all assets that can have a value estimated. Is it that:

And you wouldn't need to screw up the normal power points.

I do not see it like that, I see it as the opposite. I unscrew them and make them more comparable between sessions. This whole concept of power points is new and I am trying to replace the stupid VP system with something that better reflects what we are playing for. That you call them "normal" is quite funny. If you agree to these changes I believe the basic structure is fixed and there will only be fine tuning needed. We have the right aspects of power and we have the right parameters (year modifier, value modifier and inflation modifier) and they are all objective rather than subjective as far as possible. Well, perhaps we need to add the COTs to the basic structure, as a new aspect of power.

As for MP, you have a good point about smaller countries. A better way to handle this would be to determine how much MP can be useful based on your yearly income and the price of troops. Perhaps something like this:

Basing the average troop ratio on 7/10 infantry, and 3/10 cavalry

Yearly income / (inf costs * 7 + cav costs * 3) * 10 = useful MP.

So with 1000D yearly income, 10D inf costs, 20D cav costs, you'd have

1000D/ (10 * 7 + 20 * 3) * 10 = 62.5 . So you would plug in up to 62.5 MP for PPs.

If a country can't afford more than that, it doesn't deserve much credit for the MP it can't support. Though you could give something at least in case it gets financial support. Perhaps multiply whatever spare MP the country has by .5, or less.

Well, if you make it dependent on income you could as well scrap the whole idea of giving power points for MP, you can merely increase the impact of income.

Sadly the MP aspect is a true mess as it is since we cannot use the MP cap but have to do with current MP. E.g. in our latest save you have very few men left in your MP pool while e.g. Albania has a full MP pool. That indeed screws up the comparison between sessions.

If we would play a game with average power points rather than the merely using the end number of power points the best idea would probably to scrap giving power points for MP altogether. I wish Johan would save the displayed MP maximum value in the save. That would be so great. Then we could manipulate that number to reduce it early on for the reasons you have mentioned.

 

[King John]

No, as I have said the increase in wealth is not linear. You increase more rapidly in the beginning than at the end, probably due to a series of effects, such as
- human wars are much more common in the end phase
- you gain fewer eco tech levels
- some stop colonising

Good for you. I was talking about scenarios, and how accurately your numbers will apply in different scenarios, not the rate of wealth increase. Not that its a very important detail in any case.

Yes, one could do so. However,
1. it makes it more difficult to compare between games since your effort will be compared with the effort of your opponents, not with a more objective measurement, which is what I would like
2. I believe players prefer points rather than having them transformed into percentages, the more you transform the more far from reality you come and the more difficult for players to remember how they were made up, but I may be wrong in this
3. Much work for Tonio

If Tonio would agree to do this and themajoirty in our game would prefer it I have no serious objection.

1)There is no objectivity in comparisons between games because of different rosters and ussually different starting scenarios. However, if you want the most objective thing, you should just go with a "normal" PP system that shows true value rather than using a par system.

2)Its the difference between using a par system and a pie chart, its just different ways of translating data. What makes you think this would entail any more "transforming" than your modification idea?? And really, percentages, especially if we turned them into pie charts, are much more interesting .
3)it could just as easily amount to less work, if you left out the modification stuff.

OTOH, I'm pretty sure you and I are the only ones bothering to read these tiresome posts at this point. Which means you'll probably win these arguments by default because you're in charge(even though mine

Actually I fail to see what you have against this idea of a more accurate year modifier than the one we use for treasury size today and to use it for all assets that can have a value estimated. Is it that:

Me- "and you wouldn't need to screw up the normal Power points"

I do not see it like that, I see it as the opposite. I unscrew them and make them more comparable between sessions. This whole concept of power points is new and I am trying to replace the stupid VP system with something that better reflects what we are playing for. That you call them "normal" is quite funny. If you agree to these changes I believe the basic structure is fixed and there will only be fine tuning needed. We have the right aspects of power and we have the right parameters (year modifier, value modifier and inflation modifier) and they are all objective rather than subjective as far as possible. Well, perhaps we need to add the COTs to the basic structure, as a new aspect of power.

What I mean by "normal" is absolute values that reflect exactly what aspects of power you posess at the end of the session, as opposed to "modified". I'm glad you were able to amuse yourself with that though . I thought your general proposal for PP reform was very good- readjusting the point valuing to reward credit based on monetary worth of assets, while income carries the most weight. What would be wrong with just leaving it at that? It would be a major improvement over what we have.

I didn't know you modified treasury value. That probably isn't necessary either, because as time goes on, you get more points for having higher incomes, more tech and more other assets, thus reducing the relative value of having a set amount of ducats in your treasury. I'm sure the same principle applies to everything else, thus a modifier shouldn't be needed to balance things out once you set the values right.

Well, if you make it dependent on income you could as well scrap the whole idea of giving power points for MP, you can merely increase the impact of income.

No, I thought you understood this already. You have MP and your income determines how much of the MP you can use, so the income should be the indicator of how much of your MP you should get points for. You can't just skip the MP part, that would be lazy and would undermine the whole point.

Sadly the MP aspect is a true mess as it is since we cannot use the MP cap but have to do with current MP. E.g. in our latest save you have very few men left in your MP pool while e.g. Albania has a full MP pool. That indeed screws up the comparison between sessions.

If we would play a game with average power points rather than the merely using the end number of power points the best idea would probably to scrap giving power points for MP altogether. I wish Johan would save the displayed MP maximum value in the save. That would be so great. Then we could manipulate that number to reduce it early on for the reasons you have mentioned.

You could edit the MP values in the save before you upload it to stats. It'd be a lot of work... but perhaps you could find a few people to rotate sharing that workload. The only drawback would be that you'd have to keep the original save and then post that again with normal edits to use for leaders, and having two saves with different MP values could ruin the PP averages(but we could already have a similar problem anyway)----- Or, we could have the MP aspect seperate from the stats, and just calculate PPs for MP and add them in the PP stats that you post in the threads. This workload would also need to be divided. This would almost require you do make the average PP scores manually, but that wouldn't be too much work I would think. Either way, we wouldn't need to all peace out five years before the end of the game then.

 

[unmerged(6147)]

Good for you. I was talking about scenarios, and how accurately your numbers will apply in different scenarios, not the rate of wealth increase. Not that its a very important detail in any case.

RATE OF WEALTH INCREASE

When I made my little analysis of wealth expansion I looked at one Watk3.1 1419 start, one ****** 1492, one PE 1490 and one more which was the latest Battlefront - a ****** I think. But you said you did not mean the rate of wealth increase.

I do not think it is wise to dismiss a comparison of wealth increase rate between different scenarios just like that. Wealth is the foundation for getting power points. With wealth you "buy" manus, soldiers, fleets etc and even land.


STARTING DATE

One of the most important differences between scenarios is that their starting dates may differ - the end dates are normally the same. I see no difference in wealth increase between them although my statistical material is not that large.


ABSOLUT INCOME?

So what numbers are you referring to? The absolut income numbers? I commented upon different basic tax values and so on in my previous post. I can only spell out the obvious, of course your income will be higher if you have higher basic tax values, higher production values etc. A percentage distribution of power points, as you suggest, would eliminate such a difference.


PERCENTAGE DISTRIBUTION OF POWER POINTS

However, the more I think of it the more I am against a percentage distribution. As you remember I mentioned that this will make your number of power points depending on the performance of your opponents. Now take for example FRA in C&C2. The nation was often subbed/AI and was also badly played in general I think. Anyhow, when I compared their income after the first session with their income in the end save, some 300 years of play later, it turned out they had increased to twice as much Which is ridiculously low, normal increases are something like around 8-12 times. And we had problems with more nations in that game. ENG for example was quite lazy and did not at all grew as much as she should. Well, in such a game your nation, no doubt more well played, would get a higher percentage value than in a game where they were more skilfully played. No, I prefer a more objective system of assigning values.

1)There is no objectivity in comparisons between games because of different rosters and ussually different starting scenarios.

"There is no objectivity..." It is this kind of generalisation that draws a discussion down into the gutter. You may question the degree of objectivity but you cannot claim that the degree is zero.

Of course we have an impact from your opponents, regardless of which system you choose. Take such an easy example that if you are FRA and ENG snatches the COT in Shanghai from China AI you cannot snatch it from the AI any longer. However, a percentage system like you suggest make that impact much bigger than does a system using absolut power points only.

Example: say you play well and reach 50 power points and opponent A reaches 40 and opponent B reaches 10. Then in your world they would get these percentage numbers as well. Now assume we play one more game. You again reach 50, opponent A again reaches 40 but B reaches 40 as well, he was less AI, played more focussed on "winning", went for the right tech levels in the right order etc. Well in my world you still get 50 and the other two gets 40. In your world you would now get 38% and the other two 31%. So tell me, which system makes for a better comparison between the two games? Did your performance really drop from 50 to 38% in skill? Do note, the power system does not try to measure your part of the combined power of the human competitors, it tries to measure your level of "skill", as the name of the campaign shows.

However, if you want the most objective thing, you should just go with a "normal" PP system that shows true value rather than using a par system.

Have you become a politician? There is no substance visible in this claim. And thus I leave it as it is.

And really, percentages, especially if we turned them into pie charts, are much more interesting .

I agree that the percentage view may be interesting as well. Ideally we would have both, but I am not sure Tonio will agree to program it. You must keep in mind that all of this is possible only because Tonio freely and without charge programs it.

3)it could just as easily amount to less work, if you left out the modification stuff.

Well, he has agree to program the things I have proposed so far. TO program the value modifier and the asset modifier is not much work. The value modifier is extremly simple: it is just Value modifier = X and then you use it as a multiplier in some 5-6 places. To set the year modifier means a little more especially since Tonio agreed to set it dynamically, depending on the the exact year in the game, and not using these intervals of years I have suggested.

OTOH, I'm pretty sure you and I are the only ones bothering to read these tiresome posts at this point. Which means you'll probably win these arguments by default because you're in charge(even though mine

Hehe, I assume you intended the sentence to end "... arguments are better"

But I agree, the degree of interest our players have shown in the rules of the game has been very small. But here, as for most other rules in our game, we do not allow ourselves to change them without consensus. Only if there were some kind of emergency, a loop-hole in the rules which seriously threatened the game joy, would we be ready to disregard consensus.





What I mean by "normal" is absolute values that reflect exactly what aspects of power you posess at the end of the session, as opposed to "modified". I'm glad you were able to amuse yourself with that though . I thought your general proposal for PP reform was very good- readjusting the point valuing to reward credit based on monetary worth of assets, while income carries the most weight. What would be wrong with just leaving it at that? It would be a major improvement over what we have.

I didn't know you modified treasury value. That probably isn't necessary either, because as time goes on, you get more points for having higher incomes, more tech and more other assets, thus reducing the relative value of having a set amount of ducats in your treasury. I'm sure the same principle applies to everything else, thus a modifier shouldn't be needed to balance things out once you set the values right.






No, I thought you understood this already. You have MP and your income determines how much of the MP you can use, so the income should be the indicator of how much of your MP you should get points for. You can't just skip the MP part, that would be lazy and would undermine the whole point.



You could edit the MP values in the save before you upload it to stats. It'd be a lot of work... but perhaps you could find a few people to rotate sharing that workload. The only drawback would be that you'd have to keep the original save and then post that again with normal edits to use for leaders, and having two saves with different MP values could ruin the PP averages(but we could already have a similar problem anyway)----- Or, we could have the MP aspect seperate from the stats, and just calculate PPs for MP and add them in the PP stats that you post in the threads. This workload would also need to be divided. This would almost require you do make the average PP scores manually, but that wouldn't be too much work I would think. Either way, we wouldn't need to all peace out five years before the end of the game then.[/QUOTE]

 

[Peter Ebbesen]

If you really want more accurate manpower values (not manpower pool, which takes more effort as one has to factor in graintrading which is nearly impossible to do sensibly server-side, but actual manpower i.e. the manpower gain per year*) it can in most situations be extracted from the save files with that bit of extra effort by doing a summing over owned provinces cross referenced with province base manpower and taking into account culture, citysize, conscription centre, and revolt risk. A script to do so would be a few hours work on the outside. The only real problem would be getting the "if not same continent, check for land-connection" algorithm set up - probably too much work to be fun and certainly an easy place to make errors in any hack job, but the mechanisms are completely known and the data is all there in the save file when crossreferenced with the database, so it is certainly doable.

Good luck with your project. Sounds incredibly boring to me compared to a good old-fashioned after-game analysis of how nations performed at different times (with associated fight!) and, of course, it carries an associated risk that players who play to maximize their Power score may end up more conservative in playstyle, but it does seem like a good and civilized alternative to those who really dislike playing games without declared winners.


* Which is the really important measure anyhow.

 

[Tonioz]

land connection is what prevents any attempt to calculate real manpower indeed.

Well, i remember games without stats too, and even more, games without page 34 in 1.05 were fun a lots

But, at least stats stimulate progress of many players in different areas in rather quick way, so it ends in more quality playing in return of too much information "ready on the plate".

 

[unmerged(6147)]

If you really want more accurate manpower values (not manpower pool, which takes more effort as one has to factor in graintrading which is nearly impossible to do sensibly server-side, but actual manpower i.e. the manpower gain per year*) it can in most situations be extracted from the save files with that bit of extra effort by doing a summing over owned provinces cross referenced with province base manpower and taking into account culture, citysize, conscription centre, and revolt risk. A script to do so would be a few hours work on the outside. The only real problem would be getting the "if not same continent, check for land-connection" algorithm set up - probably too much work to be fun and certainly an easy place to make errors in any hack job, but the mechanisms are completely known and the data is all there in the save file when crossreferenced with the database, so it is certainly doable.

Good luck with your project. Sounds incredibly boring to me compared to a good old-fashioned after-game analysis of how nations performed at different times (with associated fight!) and, of course, it carries an associated risk that players who play to maximize their Power score may end up more conservative in playstyle, but it does seem like a good and civilized alternative to those who really dislike playing games without declared winners.


* Which is the really important measure anyhow.

Yes Peter. We know all the facts are in your FAQ. But as you say yourself, this is a Herculean task to calculate an exactly correct number.

Regarding its impact on the playing style (I guess you mean it would make them peaceful) I can see no such tendency. Peaceful players are still peaceful and aggressive are not. The one factor that has impact on the degree of peacefulness is our military tradition system. Take for instance myself, I am a very peaceful player. But with our MT system (which means that the more soldiers you lose in battle the better average stats will your random leaders have the next session) I am much more willing to engage in war.

Regarding MP pool and the number you get each year you are of course right. However, that number is as elusive as the cap.

When determining the MT value we use the cap and the reason is that this value is relatively easy to determine. If you control all those provinces that provide you with MP you can simply hover over the MP number on the top right of the screen and use the displayed number that says "Maximum" (at least that is the number we use, although in reality the true number may be a little higher). When you hover you also get information about the number of men you get per year, but that number may be very far from correct, and can thus not be used just like that.

Your reference to how boring it will be is a typical comment from a vet seeing a proposal from a relative him latecomer. I have experienced this a "100" times before. As you may remember this attitude was the normal one among you vets when I back then before I had entered the MP playing community made a suggestion now and then how EU MP should/could be played. The fact is, and which can easily be understood just you think about it, that there exists no conflict between a power point system and after campaign narrative analysises of what happened and how players performed. Instead they complement eachother nicely. Indeed, the existence of the power system gives even more material for a narrative writer to comment on and use as a basis for his analysises.

 

[Peter Ebbesen]

Regarding its impact on the playing style (I guess you mean it would make them peaceful) I can see no such tendency.

Ah, you misunderstand me; conservative as in more risk averse, not as in peaceful. And, of course, I might well be wrong. Still, given any point system used to measure success, it is to be assumed that the players will try to maximize their score while minimizing risk - that's nothing new, of course - but the risk does exist that you'll have some players concentrating more on the score than social interaction and, for lack of a better word, flair. Fight your wars - but try to risk less, for failures are noted and remembered in the score long after the memories of men would have consigned it to "these things happen". Is this good or bad? A bit of both, I guess.

When determining the MT value we use the cap and the reason is that this value is relatively easy to determine. If you control all those provinces that provide you with MP you can simply hover over the MP number on the top right of the screen and use the displayed number that says "Maximum" (at least that is the number we use, although in reality the true number may be a little higher). When you hover you also get information about the number of men you get per year, but that number may be very far from correct, and can thus not be used just like that.

Realistically speaking, the time to code a solution (even one taking into account the damnable land-connection issue), while much too much work compared to just taking the inaccurate current MP values for each nation noted in the file, will be much less than the time spent on recording this manually for every player for every session for every recorded game - of course, it cannot be spread out over as many individuals, but over time it will be a considerable saving.

Your reference to how boring it will be is a typical comment from a vet seeing a proposal from a relative him latecomer. I have experienced this a "100" times before. <SNIP>

Well, my apologies for setting off your self-defense-system. It appears to be a sore point with you. I said boring to ME but made no assumption whatsoever to how it would be to others (except for noticing its appeal to those who like games with declared winners) - that has nothing whatsoever to do with "new" or "old" players and everything to do with personal preference, and I wished you good luck. It will be a valuable tool to those who like it (and, indeed, it will be able to support weak players in retroactively assessing the value of their actions as regards the tangible in-game position (i.e. excluding diplomacy but measuring all other assets))

You prefer one type of scoring (precise scoring, declared winner), good for you. I prefer another (social fighting, biased national evaluations with no declared winner), good for me, and I freely acknowledge that the existence of a power system is not mutually exclusive with an after game country evaluation. (I have never stated otherwise and, indeed, given that such a system has existed all along in the form of the VP system I'd be a fool if I had, and while I'm many things, including arrogant, I'm nobody's fool )

We are not all out to get you, even though you have apparently experienced forum injustice at the hands of the "vets" hundreds of times before. We just don't all agree with you on all issues and rest assured that whenever I consider an idea, or an aspect of an idea (such as using what is fundamentally asset-counting throughout the ages to determine the winner), boring, it is because of the idea, not the person.

Anyhow, back on topic: My apologies - it really wasn't my intention to derail the conversation which was on the specifics of a new Power system (which is why I started talking about manpower), not on the usability of it, but I should have remembered the notoriously volatile nature of the MP forum and left my comments as to the usability for another discussion.

 

[Tonioz]

Daniel, do you want to calculate income growth for Portugal ?
http://www.europa2.ru/cgi-bin/leagu...ason=Valkyrie.net VIII&lang=eng&yearsave=1554

 

[unmerged(6147)]

Ah, you misunderstand me; conservative as in more risk averse, not as in peaceful. And, of course, I might well be wrong. Still, given any point system used to measure success, it is to be assumed that the players will try to maximize their score while minimizing risk - that's nothing new, of course - but the risk does exist that you'll have some players concentrating more on the score than social interaction and, for lack of a better word, flair. Fight your wars - but try to risk less, for failures are noted and remembered in the score long after the memories of men would have consigned it to "these things happen". Is this good or bad? A bit of both, I guess.

Well, this leads down to an old discussion between you and me and some other. What is the ultimate goal or our gaming? For me, as you know, it is to compete. EU for me is a game, a competition. As all games they mean a lot of fun and exciting moments but the basis is that it is a competition. And as all true competitors I appreciate when my opponents do the best they can to defeat me, because it is only then that my own victories are worth anything. So if I were to play with people who did something for "flair" or something similar that would diminish the joy I would get from the game while it apparently would enhance it for you because you consider our beloved game to be something more similar to a theater performance where the players are the actors and there is no manuscript but the actors create the action on the spot.

Realistically speaking, the time to code a solution (even one taking into account the damnable land-connection issue), while much too much work compared to just taking the inaccurate current MP values for each nation noted in the file, will be much less than the time spent on recording this manually for every player for every session for every recorded game - of course, it cannot be spread out over as many individuals, but over time it will be a considerable saving.

The math is not that easy when you are not the one who programs the code but you have to ask someone else to do it. You cannot just say to him: here spend an hour on this because it would take two hours for us to do it.

Well, my apologies for setting off your self-defense-system. It appears to be a sore point with you. I said boring to ME but made no assumption whatsoever to how it would be to others (except for noticing its appeal to those who like games with declared winners) - that has nothing whatsoever to do with "new" or "old" players and everything to do with personal preference, and I wished you good luck. It will be a valuable tool to those who like it (and, indeed, it will be able to support weak players in retroactively assessing the value of their actions as regards the tangible in-game position (i.e. excluding diplomacy but measuring all other assets))

I was not hurt or annoyed, I was merely astonished that you Peter could write something that to me indicated a clear implication that you had not understood that they can complement eachother rather than be alternatives. But apparently we agree on this.

And I did notice your "Good luck". That is why my response was so relatively friendly

 

[unmerged(6147)]

I interpret John's response as he does not veto my suggestion. And no one else have complained. Tomorrow I will ask Tonio to implement the changes I describe in post 1 including adding 2 power points for each COT owned. Unless somone else objects before that.

 

[Peter Ebbesen]

Well, this leads down to an old discussion between you and me and some other. What is the ultimate goal or our gaming? For me, as you know, it is to compete. EU for me is a game, a competition. As all games they mean a lot of fun and exciting moments but the basis is that it is a competition. And as all true competitors I appreciate when my opponents do the best they can to defeat me, because it is only then that my own victories are worth anything. So if I were to play with people who did something for "flair" or something similar that would diminish the joy I would get from the game while it apparently would enhance it for you because you consider our beloved game to be something more similar to a theater performance where the players are the actors and there is no manuscript but the actors create the action on the spot.

I see your point, but to take an example from the distant past (since I haven't played recently), I once played Russia in Machiavelli 2. It grew huge, fairly rich, and powerful. By any objective measurements based on asset counting (provinces, fortresses, armies, tech) it grew truly monstrous (stats) even when considering the huge amount of inflation taken at the end to fuel at least one player war. There was just one problem - Russia contributed next to nothing to the other players' fun, since the number of player wars and intrigue it was involved in was minimal. It reached deals with its neighbours pretty early and the neighbours were smart enough to not break the deals, and I was too focused on growing the blob to take time to break them either. It was, for all practical purposes, a SP-driven monstrous red blob in an MP world, stretching from the negotiated borders in the west and far into China and other bits of rich real estate in Asia, which accounted a lot for Russia ending up as #2 in economic VP... I just checked the stats, and it seems that your current Power system awards it generous points as well, and one based on "future power" from manufactories would probably award it an even more generous score as it was #2 in manufactories as well. Of course, I thought I did great at the time (I achieved my goals, after all), but, upon reflection, it was clear that my playing Russia that way had actually detracted from everybody's enjoyment.

Ah, well, that's water under the bridge, but as regards fun in playing and competition I'm sceptical of a scoring system that might end up excessively punishing those who actually dare to think big and throws the world into war for the sake of ambition compared to those that try to minimize expenses while performing the most cost-efficient asset buildup and perform the safest wars (aka. the best planned alliance takedowns). Once burned, twice shy, as they say. When asset-counting becomes the victory determinant, asset-preservation becomes the prime concern.


SCORING ISSUE, FORTIFICATIONS:
Interestingly enough, I notice that both the current and your prosed system is weirdly counterintuitive as regards fortifications since you use the average fortification level as the sole determinant of that score. That means that a country with few provinces but all level 4, say, will have a much higher fortification score than a country with dozens of provinces of which the most important ones are level 5+ but the rest are, say, level 1-2. This means that a player with lots of provinces out to maximize his score to win is better off spending his money on completely useless cross-realm low-grade upgrades (yeah! level 2 and 3 in the Siberian corridor! ) and not build any big fortifications unless it is absolutely necessary to his survival. This is in no way a reflection of how his "fortress strength" impacts gameplay in practise, nor is it a good measure of asset investment.

It is of course easy to calculate and it does give a good idea of how "strong" a country is in fortifications for small countries, so there's a certain virtue to that, and I understand why you don't use a simple sum over all provinces since that would make the fortress strength dominant for scoring purposes for large countries, but how about a weighted average instead based on the price increases in fortification levels? I.e. use (SUM(Ni*2^(i-1)))/nOfProvinces, i={1..6}, where Ni is the number of fortresses of level i. Or, if size does matter anyhow and one wants to add in the value of strategic depth of lots of fortified provinces anyhow without it becoming dominant, (SUM(Ni*2^(i-1)))^a, i={1..6}, where 0<a<1, tweak a to give the squishing factor needed (e.g. a=0.5 is the same as taking the square root). (Another good squishing function for sums is exp(-sum) but that's perhaps going a bit far).

 

[unmerged(6147)]

Thanks for your suggestion Peter. Although I am good at basic match I am not at more advanced.

Let me make an example to see if I understood you correctly.

Say I own 6 provinces, 1 with medium fortress, 2 with small and 3 with minimum.

Applying your basic formula this is what I get

3* 2**0 / 6
+
2 * 2**1 / 6
+
1 * 3**2 / 6

which equals
3*0/6
+
4*2/6
+
1*9/6

which equals
0
+ 8/6
+ 9/6

which is
17/6, i.e. almost 3. Was this what you meant?

Using the present formula we apply the value becomes (3*1 + 2*2 + 1*3)/6 = 10/6= 1.66.

I can see here that you give a much higher value for larger fortresses, indeed the value of a minimum fortresses will be zero since you use zero as an exponent for them. And we could see that for you two small fortresses was not even worth one medium while for me two small exceeds the impact of one medium. There are certainly differences.

BTW, the cost ratio between fortresses are
Minimal = 1
Small = 2
Medium = 3
Large = 5
Mighty = 10
Maximum = 20

These numbers could perhaps be used in some intelligent way. We could use them instead of the fortress level number (which is 1-6) when computing the dividend.

 

[Peter Ebbesen]

Thanks for your suggestion Peter. Although I am good at basic match I am not at more advanced.

It was simply that level->weight be the following:
0-->0
1-->1 (2^0)
2-->2 (2^1)
3-->4 (2^2)
4-->8 (2^3)
5-->16 (2^4)
6-->32 (2^5)

(if one went strictly by accumulated asset cost it should be 0, 1, 3, 7, 15, 31, 63, but that would perhaps be going too far - and then again, perhaps it would not, that depends entirely on what you really want to accomplish with the "fortress power value": a representation of average assets invested in each province, a representation of total assets invested in provinces, a representation of average "defensibility" of each province)

Let me make an example to see if I understood you correctly.

Say I own 6 provinces, 1 with medium fortress, 2 with small and 3 with minimum.

You got the formula wrong. With my suggestion the value would be (3*1 + 2*2 + 1*4)/6 ~ 3.67, not (3*0 + 4*2 + 1*9)/6 [for some reason you took the power of Ni rather than 2 ]

If all the provinces were small, the value would be (6*2)/6 = 2.0
If three provinces were small, three minimum the value would be (3*1 + 3*2)/6 = 1.5
If all three provinces were minimum it would be (6*1)/6 = 1.0