In another thread Tonio and I discussed a potential ”power” value algorithm to present e.g. on the stats page. A value would be presented that tried to portray the actual and potential power (economy/military) of each nation. Tonio asked to come up with a formula. I immediately started thinking and this is the preliminary result of that. As this response does not have anything to do with that particular game thread I start a new thread.
OVERVIEW
The following questions must be analysed.
A. On what features shall we base the power value?
B. Detailed analysis for each feature
C. Other
Aim: We should try and determine how to measure the power value of each feature and then also determine their relative importance – i.e. assigning some kind of weight value – e.g. a multiplier.
A. ON WHAT FEATURES SHALL WE BASE THE POWER VALUE?
Right now I can come up with the following candidates to base the power value upon.
1. Income
2. The MP pool cap
3. Army and navy morale
4. Techs
5. Ship support
6. Number of War ships and galleys
7. Forts
8. Army Size
9. Shipyards
10. Manu
11. CCs
12. Sliders
I have tried to order them according to my own view of importance. I.e. I consider income the most important feature.
Concerning sliders there are good and there are bad slider positions for sure. But I do not think they should have any influence on the power system, for the simple fact that the effect of the sliders are reflected in other features, mainly income and army morale. Thus I will ignore them.
B. DETAILED ANALYSIS OF EACH FEATURE
B1. Income:
The income is easy to obtain from the save. We should deduct income from peace and events since they do not represent lasting values. Then we should divide that income with 1+inflation (as that value is stored in the save, i.e. in the save 17% inflation is stored as 0.17).
Weight: We could e.g. set each remaining 300d in yearly income = 1 "power" point.
B2. MP pool cap.
This value must be manually fed to the program. It is much too difficult to calculate it upon stored data I think.
Weigth: Each 25,000 men in the MP pool cap could be worth 1 power point.
Note: this power value will be relative small in comparison to the value of the other features before CCs - not at all the 2nd most important. But after CCs become available it will sky rocket.
B3. Army and navy Morale.
The stats site presently gives a theoretical max value for morale (at 100% maintenance). That can be used.
I do not think it is necessary to adjust the power from navy morale for the fact that nations with low NT often have no fleet at all because the CRT level initself influences morale.
Weight: For each 0.5 in morale you get one point. Once for army and once for navy. Both values capped at 7 of course.
B4. Techs.
For each infra/trade tech level you get 6 ”tech points”.
For each LT/NT you get 1 tech point.
Some tech levels are more valueable than other (e.g. expensive levels and CRT levels) but it appears difficult to adjust to that. A bonus for these levels could however be hardcoded into the program as they are known.
Weight: Divide the added tech points (as computed according to above) by 10 and then set one remaining tech point = 1 power point
B5. Ship support
This value must also be manually fed to the program.
Since ships are much more expensive than soldiers it has more impact on the actual power. Even in the end game no one can easily build up to their ship support limit from scratch just like that but they can all do it with their MP.
Another factor is that while nations are mostly on a similar LT level there can be huge differencies for NT levels. A high ship support for Russia can e.g. be almost without any power value because they are several CRT levels behind the tech leaders. Thus to compute this value is complicated.
Weight: You could get one support point for each 100 ships supported. Then for each naval CRT you are behind the tech leader you multiply the base number with say 2. That is, if the tech leader is for example 2 CRT levels ahead of you then you get one point for every 400 ships supported (instead of 1 per 10).
B6. Number of War ships and galleys
When time passes on galleys become less useful. Also your ships become less useful if you are CRT levels behind your enemy.
Weight: Give 1 point for each 100 war ships and 1 for each 100 galleys.
For each NT CRT you are behind the tech leader multiply base value with 2.
For galleys also: If tech leader is at NT17 multiply base value for galleys with 1.5, if at NT21 multiply by 2, if at NT26 multiply by 3 if at NT if at 31 multiply by 4, if at 41 multiply with 5.
B7. Forts.
Here we have two alternatives. We could go for average fort level (already in the stats) or adding absolute numbers for forts (say 1 for each minimal, 2 for each small etc to get a total for ”fort points”). I guess average fort level is the much more important when we speak of power. Average fort levels are normally 2 point something when the game ends, i.e. on average small to medium forts.
Weight: One power point for each 0.4 average fort level.
B8. Army size.
This should be measured in army units, i.e. 10 artillery counts as 1000 men infantry/cavalry. Since it is so comparatively easy to rebuild a small army this has little impact on power. The land military strength is more accurately portrayed by MP pool and morale.
Weight: for each 200 units one power point.
B9. Shipyards
Shipyards are quite useful if you are a naval power, else mainly only useful as giving you a colonist extra.
Weight: For each 2 shipyards you get 1 power point.
B10. Manus
An important benefit of a goods manu or a refinery it their contribution to the TE/PE. However, that is covered by the income feature. What remains is thus mainly the contribution to tech investment. This makes manus very good in the beginning of the game. But almost useless very late in the game.
Also some manus are more important than other. Refineries are by far the best type (as far as you have or at least will have a good income from trade). There are too many variables involved here and a simplified approach must be applied.
Thus manus more than most other things on this list represents potential rather than acutal power. They definitively earn their place on the list in the early game and clearly carry almost no value at all on their own in say 1819.
Weight: Each refinery give 1p. Each 2 weapon manus give 1 point. The rest are added together and you get 1 for each 3 you have. If the current year is later than 1500 multiply base value with 1.5. If current year us 1600 multiply by 2. If current year is 1700 multiply with 3. If current year is 1800 multiply with 5.
B11. CCs
Do note that the importance of CCs is not so much the increased build capacity you get from them but the extra MP they provide and the latter is reflected in the MP pool cap value.
Weight: For each 10 CCs one power point.
D OTHER
If you agree with me that you ”win” a EU MP game if you are the strongest nation at the end, then this algorithm, if successfully implemented, could provide a basis for deciding who that will be. As some of you know it is a favourite theme of mine to try and establish a kind of objective system to judge the performance – as opposed to the subjective system used e.g. in DU 1.5 (not to speak about the current VP system that is essentially broken as the points distributed have little to do with skill in play).
If we want to use it to determine a winner we must weight the final number of power points according to which nation it is. Assume e.g. POR and FRA ends up with the same amount of points, then clearly POR has performed better than FRA.
We need a multiplier to fix this. I.e. we take the final number of power points and for each nation we multiply that number with that nation’s own multiplier.
Depending on scenario nations have better or worse potentials. Portugal e.g. has far better potential in Ryo’s AoD on the new map than in vanilla. A campaign starting in 1419 further more gives POR an even better position. As does maps with more gold out there to catch early on.
As can be seen a lot depends on the scenario. Thus the simplest thing is for each campaign to define their own multipliers for nations. In fact the nations can be auctioned away based upon this multiplier value, the one giving the lowest multiplier gets the nation.
As for default multipliers in vanilla 1419 the following values could perhaps be reasonable.
FRA 1
OE 1
SPA 1.1
ENG 1.1
RUS 1.2
POR 1.3
NL 1.3
SWE 1.5 (assuming no DEN)
AUS 1.7
DEN 2 (assuming no SWE)
BB 2
VEN 2
POL 2.5
And of course, if both SWE and DEN are in the game then their multipliers should be higher than suggested above. Similar considerations can be made on other parts of the maps but the Scandinavian situation is perhaps the most obvious one since the one existing will almost always eat up all of the other one.
A POSSIBLE EXAMPLE
For a really powerful nation in 1819 I predict something like this
1. Income (adjusted for inflation) : 9000 = 30 power points (PP)
2. MP pool cap: 700 = 28 PP
3. Morale: 6.5 in land, 6 at Sea = 25 PP
4. Techs: 60+60+9+8=222 = 22 PP
5. Ship support: 1200 = 12 PP
6. Number of warships and galleys: 1000 war ships: 10 PP
7. Forts: average level 2.5 = 6 PP
8. Army size: 1.000.000 men = 5 PP
9. Shipyards: 8 = 4 PP
10. Manus: 15 refineries, 10 weapon, 3 other: 3+1= 4 PP
11. CC 20 = 2 PP
ANOTHER POWER SYSTEM
Incidentally there already exists a kind of power system. AFAIK it was created by Zeitgeist and used succesfully in the Thirst of Glory campaign I and now also TfG II. But the scope of that system was not the same as described above. The TfG power system is smaller to its extent AFAIK and serves as a basis for rewards for the nextcoming sessions (the more powerful you are the more rewards you get).
FINAL WORDS
Remember, as all values in this analysis to a very large extent constitutes pure guess work I do not claim they are well chosen. They are merely suggestions.
Now let us hear from Tonio if he thinks he could program this for us and present it oin e.g. his normal stats site?
Everyone else if of course also invited to contribute to this idea.
-----
Coming to think of it, if we are reasonably successful in describing a system of this type we could post it in the EU3 thread. Who knows
OVERVIEW
The following questions must be analysed.
A. On what features shall we base the power value?
B. Detailed analysis for each feature
C. Other
Aim: We should try and determine how to measure the power value of each feature and then also determine their relative importance – i.e. assigning some kind of weight value – e.g. a multiplier.
A. ON WHAT FEATURES SHALL WE BASE THE POWER VALUE?
Right now I can come up with the following candidates to base the power value upon.
1. Income
2. The MP pool cap
3. Army and navy morale
4. Techs
5. Ship support
6. Number of War ships and galleys
7. Forts
8. Army Size
9. Shipyards
10. Manu
11. CCs
12. Sliders
I have tried to order them according to my own view of importance. I.e. I consider income the most important feature.
Concerning sliders there are good and there are bad slider positions for sure. But I do not think they should have any influence on the power system, for the simple fact that the effect of the sliders are reflected in other features, mainly income and army morale. Thus I will ignore them.
B. DETAILED ANALYSIS OF EACH FEATURE
B1. Income:
The income is easy to obtain from the save. We should deduct income from peace and events since they do not represent lasting values. Then we should divide that income with 1+inflation (as that value is stored in the save, i.e. in the save 17% inflation is stored as 0.17).
Weight: We could e.g. set each remaining 300d in yearly income = 1 "power" point.
B2. MP pool cap.
This value must be manually fed to the program. It is much too difficult to calculate it upon stored data I think.
Weigth: Each 25,000 men in the MP pool cap could be worth 1 power point.
Note: this power value will be relative small in comparison to the value of the other features before CCs - not at all the 2nd most important. But after CCs become available it will sky rocket.
B3. Army and navy Morale.
The stats site presently gives a theoretical max value for morale (at 100% maintenance). That can be used.
I do not think it is necessary to adjust the power from navy morale for the fact that nations with low NT often have no fleet at all because the CRT level initself influences morale.
Weight: For each 0.5 in morale you get one point. Once for army and once for navy. Both values capped at 7 of course.
B4. Techs.
For each infra/trade tech level you get 6 ”tech points”.
For each LT/NT you get 1 tech point.
Some tech levels are more valueable than other (e.g. expensive levels and CRT levels) but it appears difficult to adjust to that. A bonus for these levels could however be hardcoded into the program as they are known.
Weight: Divide the added tech points (as computed according to above) by 10 and then set one remaining tech point = 1 power point
B5. Ship support
This value must also be manually fed to the program.
Since ships are much more expensive than soldiers it has more impact on the actual power. Even in the end game no one can easily build up to their ship support limit from scratch just like that but they can all do it with their MP.
Another factor is that while nations are mostly on a similar LT level there can be huge differencies for NT levels. A high ship support for Russia can e.g. be almost without any power value because they are several CRT levels behind the tech leaders. Thus to compute this value is complicated.
Weight: You could get one support point for each 100 ships supported. Then for each naval CRT you are behind the tech leader you multiply the base number with say 2. That is, if the tech leader is for example 2 CRT levels ahead of you then you get one point for every 400 ships supported (instead of 1 per 10).
B6. Number of War ships and galleys
When time passes on galleys become less useful. Also your ships become less useful if you are CRT levels behind your enemy.
Weight: Give 1 point for each 100 war ships and 1 for each 100 galleys.
For each NT CRT you are behind the tech leader multiply base value with 2.
For galleys also: If tech leader is at NT17 multiply base value for galleys with 1.5, if at NT21 multiply by 2, if at NT26 multiply by 3 if at NT if at 31 multiply by 4, if at 41 multiply with 5.
B7. Forts.
Here we have two alternatives. We could go for average fort level (already in the stats) or adding absolute numbers for forts (say 1 for each minimal, 2 for each small etc to get a total for ”fort points”). I guess average fort level is the much more important when we speak of power. Average fort levels are normally 2 point something when the game ends, i.e. on average small to medium forts.
Weight: One power point for each 0.4 average fort level.
B8. Army size.
This should be measured in army units, i.e. 10 artillery counts as 1000 men infantry/cavalry. Since it is so comparatively easy to rebuild a small army this has little impact on power. The land military strength is more accurately portrayed by MP pool and morale.
Weight: for each 200 units one power point.
B9. Shipyards
Shipyards are quite useful if you are a naval power, else mainly only useful as giving you a colonist extra.
Weight: For each 2 shipyards you get 1 power point.
B10. Manus
An important benefit of a goods manu or a refinery it their contribution to the TE/PE. However, that is covered by the income feature. What remains is thus mainly the contribution to tech investment. This makes manus very good in the beginning of the game. But almost useless very late in the game.
Also some manus are more important than other. Refineries are by far the best type (as far as you have or at least will have a good income from trade). There are too many variables involved here and a simplified approach must be applied.
Thus manus more than most other things on this list represents potential rather than acutal power. They definitively earn their place on the list in the early game and clearly carry almost no value at all on their own in say 1819.
Weight: Each refinery give 1p. Each 2 weapon manus give 1 point. The rest are added together and you get 1 for each 3 you have. If the current year is later than 1500 multiply base value with 1.5. If current year us 1600 multiply by 2. If current year is 1700 multiply with 3. If current year is 1800 multiply with 5.
B11. CCs
Do note that the importance of CCs is not so much the increased build capacity you get from them but the extra MP they provide and the latter is reflected in the MP pool cap value.
Weight: For each 10 CCs one power point.
D OTHER
If you agree with me that you ”win” a EU MP game if you are the strongest nation at the end, then this algorithm, if successfully implemented, could provide a basis for deciding who that will be. As some of you know it is a favourite theme of mine to try and establish a kind of objective system to judge the performance – as opposed to the subjective system used e.g. in DU 1.5 (not to speak about the current VP system that is essentially broken as the points distributed have little to do with skill in play).
If we want to use it to determine a winner we must weight the final number of power points according to which nation it is. Assume e.g. POR and FRA ends up with the same amount of points, then clearly POR has performed better than FRA.
We need a multiplier to fix this. I.e. we take the final number of power points and for each nation we multiply that number with that nation’s own multiplier.
Depending on scenario nations have better or worse potentials. Portugal e.g. has far better potential in Ryo’s AoD on the new map than in vanilla. A campaign starting in 1419 further more gives POR an even better position. As does maps with more gold out there to catch early on.
As can be seen a lot depends on the scenario. Thus the simplest thing is for each campaign to define their own multipliers for nations. In fact the nations can be auctioned away based upon this multiplier value, the one giving the lowest multiplier gets the nation.
As for default multipliers in vanilla 1419 the following values could perhaps be reasonable.
FRA 1
OE 1
SPA 1.1
ENG 1.1
RUS 1.2
POR 1.3
NL 1.3
SWE 1.5 (assuming no DEN)
AUS 1.7
DEN 2 (assuming no SWE)
BB 2
VEN 2
POL 2.5
And of course, if both SWE and DEN are in the game then their multipliers should be higher than suggested above. Similar considerations can be made on other parts of the maps but the Scandinavian situation is perhaps the most obvious one since the one existing will almost always eat up all of the other one.
A POSSIBLE EXAMPLE
For a really powerful nation in 1819 I predict something like this
1. Income (adjusted for inflation) : 9000 = 30 power points (PP)
2. MP pool cap: 700 = 28 PP
3. Morale: 6.5 in land, 6 at Sea = 25 PP
4. Techs: 60+60+9+8=222 = 22 PP
5. Ship support: 1200 = 12 PP
6. Number of warships and galleys: 1000 war ships: 10 PP
7. Forts: average level 2.5 = 6 PP
8. Army size: 1.000.000 men = 5 PP
9. Shipyards: 8 = 4 PP
10. Manus: 15 refineries, 10 weapon, 3 other: 3+1= 4 PP
11. CC 20 = 2 PP
ANOTHER POWER SYSTEM
Incidentally there already exists a kind of power system. AFAIK it was created by Zeitgeist and used succesfully in the Thirst of Glory campaign I and now also TfG II. But the scope of that system was not the same as described above. The TfG power system is smaller to its extent AFAIK and serves as a basis for rewards for the nextcoming sessions (the more powerful you are the more rewards you get).
FINAL WORDS
Remember, as all values in this analysis to a very large extent constitutes pure guess work I do not claim they are well chosen. They are merely suggestions.
Now let us hear from Tonio if he thinks he could program this for us and present it oin e.g. his normal stats site?
Everyone else if of course also invited to contribute to this idea.
-----
Coming to think of it, if we are reasonably successful in describing a system of this type we could post it in the EU3 thread. Who knows
Last edited: