To Kill or Be Killed: That Is The Question
By Chris Brantley
What is the most effective troop type in DBA? Is it Blades? Possibly Knights or Pikes? I'm sure everyone has their own opinion. It may depend on your tactics. It may depend on your psychological disposition. It may even depend on your die rolling. But surely there must be some way to quantify who stands at the top of the heap.
Frank Alejano took a stab at it with his posting to the rec.games.miniatures.historical newsgroup of the following table of kill probabilities in DBA. It doesn't cover every troop type to be sure (someone else will have to figure the relative merits of Elephants and War Wagons). But it does provide some interesting grist for the mill. The numbers in the chart represent how many chances out of 36 possible results that an element listed in the left column will kill a corresponding element on the right axis either by quick kill or doubling total. I did not have the patience to check Frank's math, but must leave it to others to verify the data underlying the theorum.
Chance of Killing (x of 36) 
Unit 
Spears 
Blades 
Cavalry 
Knights 
Warband 
Auxilia 
Pikes 
2xPikes 
Spear 
1 
0 
3 
3 
3 
3 
3 
0 
Blades 
2 
0 
2 
2 
6 
6 
2 
0 
Cavalry 
0 
0 
2 
0 
6 
6 
0 
0 
Knight 
10 
2 
3 
1 
21 
21 
10 
3 
Warband 
10 
5 
1 
1 
2 
2 
15 
3 
Auxilia 
0 
0 
1 
1 
2 
2 
2 
0 
Pikes 
0 
0 
3 
3 
2 
2 
2 
0 
2xPike 
4 
1 
12 
12 
9 
9 
9 
0 
So what do we make of this data? Leaving out supported Pikes and their +3 bonus for a moment, one measure of relative effectiveness would be to compare each troop types' total chances of killing other elements in an encounter against the likelihood of being killed, the result being what I will refer to as a Mortality Ratio (MR). In addition, if you total the number of chances that an element will kill or be killed and divided the result by the total number of possibilities given the adversaries listed (7x36=252) then you have a nice measure of how lethal the combat pairing is likely to be. The higher this Lethality Ratio (LR), the more decisive the outcome of the close combat.
Unit 
To Kill 
Or Be Killed 
Mortality Ratio 
Lethality Ratio

Spear 
16 
23 
0.70 
15.5% 
Blades 
20 
7 
2.86 
10.7% 
Cavalry 
14 
15 
0.93 
11.5% 
Knights 
68 
11 
6.19 
31.3% 
Warband 
36 
42 
0.86 
30.9% 
Auxilia 
8 
42 
0.19 
19.8% 
Pikes 
12 
34 
0.35 
18.2% 
And what conclusions can we draw from this data? First that Knights (MR 6.19) are killing machines, followed at a respectable distance by Blades (MR 2.86). Also that Auxilia (MR 0.19) are the wimps that everyone knows them to be. And perhaps surprisingly that Pikes (MR 0.35) are remarkably suspect, falling significantly behind Cavalry (MR 0.93), Warband (MR 0.86), and Spear (MR 0.70) in relative mortality. Second, if you like troops that are tough to kill, Blades are the troop type for you with only a 3% of succumbing across the board (7 death results out of 252 chances). Combining it's relatively high Mortality Ratio with its outstanding survivability probably reenforces claims that Blades are the most "effective" DBA element. But then again, the Knights look particularly strong. And finally, if you like your battles short and to the point, then the impetuous Knights (LR 31.3%) or Warband (LR 30.9%) are your best prospects. I would caution the gentle reader, however, to take these conclusions with a grain of salt. After all there are lies, damned lies, and then there are statistics. Some of the ratios are inflated due to particularly onesided mismatches. If you remove the Knights versus Warband or Auxilia matchups from the table, for example, the ratios would shift significantly. Similarly, if you take into account the missing troop elements (i.e. Psiloi, Scythed Chariots, Elephants, Light Horse, Bows, and War Wagons), the totals and resulting ratios might look completely different. Finally, no attempt is made to weigh the effect of support bonuses or bad going modifiers. Despite these cautions, the numbers do test your assumptions about the relative merits of the DBA troops types and give some pause to ponder. And if you fervently disagree with the numbers or my conclusions, (for example, you may feel that Spear is not given a fair shake) that is okay...for mine tis but a tale told by an idiot, full of sound and fury, signifying nothing. On that note, I must close this essay to let the dog out for his nightly excursion. Out damned Spot. Comments, suggested additions, and/or critiques welcome. Direct them to Chris Brantley at brant@erols.com.
