Determining a Field Army's Winning Percentage!

[Joe Meyer]

Gentlemen of the Union Army,

I have been questioned on the method of determining the overall winning percentage in regards to the monthly Win-Draw-Loss records of the various field armies, a subject that I confess has always caused me some unease. So I have decided to put the question right on out there to all of you to see what formula you might like to see and adopt. We are speaking here of the formula that would be used to determine the UA's Quarterly Combat Medallion which is awarded to that UA field army turning in the best combat performance for each quarter of the year.

The current formula that I use totals all of the Major and Minor victories and divides that by the total number of games played, including DRAWS!

Is this a fair formula? Should there be minimums to be met before being computed? Should the formula be changed?

Give me your thoughts on this, and when we are finished discussing it I will arrange a poll vote for any and all alternate formulas so that we may make a final determination!

[cruces]

My formula would be different-

Major Victory +3 points
Minor Victory +2 point
Draw +1
Minor Loss -1
Major Loss -2

Total times games played

highest Army winner takes the medal. Rewards for game play but factors in losses.

MG Elkin
3rd Div/ (2nd Cav) XVIth Corps
AotT

[krmiller_usa]

I use a similar formula to Gen. Elkins for determining the winner of the quarterly Sheridan Combat Medal and something on that order might be more appropriate for the Army Combat Medallion rather than a straight percentage. However when reporting the a winning percentage you cannot count draws it should be strictly victories/ victories+defeats, that was the method used in football before they had overtimes.

[Dirk Gross]

However when reporting the a winning percentage you cannot count draws it should be strictly victories/ victories+defeats, that was the method used in football before they had overtimes.

I agree about draws. I think they should not be included in the calculation.

[Dwight McBride]

I agree with General Myers's original idea: divide games won by all games played (including draws).

I know that's a little 'hard ass,' but since we're after how often a player actually WINS a match, I feel it important to include all games NOT WON, which includes draws.

[Douglas E Shaw]

My proposal would be MJV = 4, MNV = 3, Draw = 2, MNL = 1 and MJL = 0. You then multiply the factor by the number of games in each category and add the total together. Next divide this total by 4 times the total number of games. This gives you a percentage score between 1.000 and 0.000 like a batting average. A 1.000 average equates to all major victories, a 0.750 = all minor victories, a 0.500 = all draws, a 0.250 = all minor losses and a 0.000 equals all major losses. You can tell easily whether an Army is doing better than average or below average (0.500 being average). I do agree that you have to count draws.

[ernie sands]

I think we will get a variety (as we already have) of formulas for this.

The point is to determine the overall WIN percentage. So, draws, minor losses and major losses should NOT enter in to the formula.

So, IMO, the formula should be much simpler. Major Victories plus Minor Victories= total victories.

Then do that for each army. Then add that together for a total and divide.

For example: Army 1 = 10 total
Army 2 =17
Army 3 = 7
Army 4 = 21

10+17+7+21=55

Divide each Army total by total games and you get A1=18%;A2=31%;A3=13%; A4=38%

If you are talking about a percentage of TOTAL games played, then the calculation would be similar, except you would include draws and losses.

[PAT]

that started this discussion:

I've had a small quarrel for some time with the way you present your engagement winning percentage.
I think that a draw should be considered a neutral, rather than a loss.

For Example: My AotS had a record of 9 MV, 5 mV, 10 draws, 2 mD, and 6 MD.
To me that looks like 14 wins and 8 losses, surely a winning record!! 64%... Along with those 10 draws, which are neither a win nor a loss.
The AotC would be 12 - 2, a great record. 86%
The AotT would be 3-0, undefeated, but is anyone playing?? 100%
The AotP went 4-7, our only losing record. 36%

So the Union went 33-17... looks pretty fantastic to me!! Winning almost 2 of every 3 decisions, 66%... with the 17 draws being neutral.

Our side kicked some butt big time!! But your numbers make us look like we are on the losing end at 49%.

[tommy.lynn]

I totally agree that draws should not count as a loss,a lot of people are commiting a lot of gaming time, and at the time mentioned, I was serving in VIII Corps AotS and I felt as though we had been kicked in the slats using the current formula.

[elpaco]

Your formula appears wrong to me Joe :

let me take a simple example :
Army A record : 10-0-20-0-0 (major victory-minor vict.-draw-minor defeat-major def.)
Army B record : 10-0-0-0-20
According to your formula, both armies are equal with a 33% winning percentage
According to me, I would rather give the award to Army A who had NO loss !!!

another example, even worse :
Army A record : 20-0-80-0-0 : your record : 20%
Army B record : 10-0-0-0-20 : your record : 33%
According to your formula, Army B would have the record although it scored less wins and 20 major defeats to none to army A !!!

I would rather suggest to use something like MG Elkin has in mind or do as General Miller proposes : use the same as for determining the Sheridan badge (and that would also add consistency on our different calculations) :
Maj V : +2
min v : +1
draw : 0
min d : -1
Maj D : -2



To Ernie : I think we can't just add the number of victories, as there would be a strong advantage to the army that played the higher number of games.

About the minimum number of games, I would suggest to add something like : an army can only compete if it has completed 20 games in a month (you'll have to check if 20 is realistic according to your archives). Otherwise, it would be possible to have an army with only one major victory (and no other games) clinch the title...