Jerman and I spent some time recently analyzing our respective pools of tourney entries and I thought I'd pass along a few things -- some surprising to me --that I learned from mine. I'll defer to Jerman on his. My data represents most of my tourney entries since I returned in July of 2014. It includes 895 entered warriors and 74 "successes", the latter being defined as having reached round 10.
My hope is that this will help others who are still fighting the good fight to design warriors that will do well in tourneys. Perhaps primarily by what hasn't worked for me. I suspect none of the Megas will be surprised by my revelations.
A word about the statistics: If you're going to cry "sample size" or "what's the two-sided confidence interval margin?", this is probably not the post for you. Some of these designs would be counted multiple times for the same warrior, e.g. if I entered a warrior in rooks and then apps. So things like knack could impact the numbers. The analysis does not take into account bonuses in physicals or skills, strategy, or really anything but base design and result. It doesn't take into account the primary physicals, i.e. ST/CN. Finally, I'm only just getting an ADM crew stood up so this analysis only has success in basic.
Finally, consider the source here. This is how well I have done and my managerial skills are dubious (if getting better). Any of these are sort of like saying "Stallone's funniest movie yet!"
1. I've entered 71 warriors with 17 WT/WL across nine styles (no PRs)and none of them made it to round 10. Zero. That includes a couple ABs that had 21 DF to boot, but I consider that an anomaly.
2. I've entered 53 warriors with a 15 WT and either a 17 or 21 WL across seven styles (excepting PR, PS and ST) and none made it to round ten.
3. There are six WT/WL/DF designs that have accounted for 36% of my warriors making it to round 10. Six. Out of 274 unique WT/WL/DF designs entered. Each of these designs achieved a 20% or greater success rate with a minimum of three successes. I.e., I did not count a design that achieved only 1 or two successes as being too susceptible to introducing bias and muddying the waters.
Two of them only have a total of 8 entries each or, as my buddy who testifies for me when I need statistical expertise would say, "observations". Hence the "sample size" disclaimer above. The other four designs average 20+ observations, i.e. entries in a tourney.
a. Let's get the "no shit, Sherlock" one out of the way: I've achieved success with 21 WT / 21 WL / 11 DF LUs on 28% of the entries. Megas, I hope you're jotting this down.
b. Since we're on LUs, 17 WT / 21 WL / 11 DF has gotten to round ten 22% of the time. Again, not too surprising. I assume we're all running these when we can.
c. A PR design of 17 WT / 8 WL / 17 DF has succeeded 37.5% of the time. This is one of the two eight-entry observations.
d. Total parrys with 5 WT / 21 WL / 11 DF have made it round 10 37.5% of the time as well. This is the other small-entry result.
e. STs with a 21 WT / 9 WL / 11 DF design achieved a 20.83% success. If anything cries out for an easy design with a good chance of success, it's this one.
f. Finally, ABs with a 15 WT / 15 WL / 21 DF have been successful 20% of the time. This is interesting for me because it is more successful than I have been with better rollups, e.g. 15 WT / 17 WL or 17 WT / 17 WL.
None of those 6 designs are shocking. Again, it's more about what designs were consistently not successful that I found most interesting. Your mileage will vary. I know some folks, Floyd for example, has a couple favorite 17 WT/WL tourney designs.
My plan for January (other than to run all STs) is to basically whittle out all the 17/17 and 15/21 warriors I had bucketed as potential entrants and see if that improves my performance.
Happy to share other data points if anyone would find that interesting.
You cannot post new topics in this forum You cannot reply to topics in this forum You cannot edit your posts in this forum You cannot delete your posts in this forum You cannot vote in polls in this forum