Cool, time to explain some elementary statistics to some very egotistical minds. Lucky for us, we have nice values to work with. Starmie hits Tauros for between 163 ~ 192, with an average of 177.5, which is 50.3%. Round down to 50% (which is AGAINST starmie btw) for the sake of insanity. 50% is an awesome number. Now to solve using the simplest method (aka the longest) such that even the most elementary minds could understand.
Odds of Starmie KOing on turn 1: .225 * .8 * .5 = 9%
- high roll CH hit only
Odds of Starmie KOing on turn 2: .8 * .775 * .8 + .8 * .225 * .5 * .8 + .2 * .225 * .8 * .5 = 58.6%
- First hit regular non-CH, second hit anything; first hit low roll CH, second hit anything; first hit miss, second hit high roll CH
Odds of Starmie KOing on turn 3: 2 * (.8 * .775 * .2 * .8) + 2 * (.8 * .225 * .5 * .2 * .8) + .2 * .2 * .8 * .225 = 23.44%
- First hit regular <-> second hit miss, third hit anything; first hit low-mid roll CH <-> second hit miss, third hit anything; first two hits miss, third hit high CH
Odds of Starmie KOing by turn 3: 9 + 58.6 + 23.44 = 91.04% (aka before a for sure death)
To clarify, this means there's a ~9% chance of either missing 3 (0.8%) in a row, or missing 2 and the one hit isn't a CH. Let's face it, once it reaches this Stage, Starmie is probably dead.
For all intents and purposes, there's a 8.9% chance to KO by turn 4 (it diminshes to ridculously insignificant values after it anyway).
Now this doesn't really mean anything, let's take a look at the weighed number of hits it'll take to down Tauros, or expectation.
E[X]: 1*(.09) + 2*(.586) + 3*(.2344) + 4*(.089) =
2.32HKO
Alakazam hits Tauros for 151 ~ 178, which means only about a 9.5% chance to deal "significant" damage. Let's round it up to 10% (in Alakazam's favor btw). Unlike with starmie, where the average is 50%, meaning if we consider an even number of attacks, on average, things balance out to 50%, meaning they cancel out. However, we still calculate low/high rolls for CHs, because it results in an odd number of attack. With Alakazam, things get a little bit more confusing, as we have to differentiate each attack into low or high roll values. That leaves us with 4 kinds of attacks we have to consider, low roll/high roll CH/non CH.
Odds of Alakazam KOing on turn 1: .235 * .1 = 2.35%
- high roll CH only
Odds of Alakazam KOing on turn 2: .9 * .235 + .235 + .1 * .1 = 45.65%
- First hit low roll CH, second hit anything; first hit regular anything, second hit CH anything; first hit high roll regular, second hit high roll regular
Odds of Zam KOing before a for sure death: 2.35 + 45.65 = 48%
Alakazam would be dead here. But let's continue, for fun:
Odds of Alakazam KOing on turn 3: 100 - 45.65 - 2.35 = 52%
- Whatever's left
E[X]: 1*(.0235) + 2*(.4565) + 3*(.52) =
2.5HKO
On the other side of the boat, let's look at some more facts:
Both outspeed Tauros
Average 51% against Zam with Body Slam
Average 35% against Starmie with Body Slam
Average 89% against Zam with HB
Average 61% against Starmie with HB
There are no attacks in OU (or BL for that matter) that averages less than 11% to Alakazam (Slowbro's Psychic, Persian's Bubblebeam).
Actually, that's a lie, HB does 279.5 on average, and BB does 32.5, for a total of 312, just shy of Zam's 313.
So I guess Zam is better against Tauros. Y'know, because of opinions and stuff like that or something. Yeah, it's not.
I know HP was never popular in RBY's prime, or GSC's for that matter, perhaps it's time to get out of that shell?
No one KOs Chansey with zam unless the game's already over. 10 psychics for -3 fall? nothx
(i.e. Nobody uses Golem and Rhydon together, or Starmie and Lapras.)
I'm not that biased. Lapras goes to Zam, unless it uses Body Slam or something.
Note: Any doubts about my numbers (because I have them, but I know they're close if there are errors), here are some guesstimates
Starmie odds of landing 2 consecutive HP: 64%
64 - 9 = 55 which is roughly equal to 58.6, but less than because of CH possibilities
Odds of Alakazam CHing on at least one turn by turn 2: 2 * .235 = 47%
47 - 2.35 = 44.65 which is roughly equal to 45.65, but less than because of CH possibilities