User blog:BunsenH/Ghazt stats 3

I've been a bit too busy to post an update lately, but here we go. The "minions" and I passed 4000 Gazt breeding attempts recently, so here are the latest stats. Our procedures are the same as before: the T-Rox and Entbrat both at Level 4 and 0% happiness, with those monsters immediately beside each other.

Of the 4140 tries, 42 have produced Ghazts. This gives a success probability of (1.01 ± 0.16)%. This is a bit larger than the values at our previous "milestones". And the uncertainty is about half what it was after just 950 attempts, as expected given the 1 / √ N  relationship.

The two sub-groups are the same as before. In the first, in which the player level is pushed as high as possible by XP from food production, the levels are now all 21 or 22. This group has 2415 breeding tries and 21 Ghazts bred, a probability of success of (0.87 ± 0.19)%. This is a little higher than the previous value and as expected, the uncertainty is lower. In the other group, the player levels are kept at 11, the lowest possible for breeding a Ghazt. This group now has 1886 breeding tries and likewise 21 Ghazts bred, a success probability of (1.22 ± 0.26)%. That's down from before, but again within the limits of uncertainty, especially given the small number of tries in this subgroup before. Note that the uncertainty value is much smaller than before, mostly because of the increase in the number of tries.

I haven't decided how long to continue this experiment. The "minions" are willing, given consideration for some slack time over the summer. The probability values for the two subgroups are converging, but still quite distinct, and I don't know what to make of that.