Q2: "If you have 5 #30 scores [out of 250 games recorded], should you get in? How about 5 #40’s? #50’s?
It turns out that exactly the top 20% of participants in A division made the playoffs this year. Assuming I didn’t screw up a calculation here: if a person had a ticket with exactly a top 20% score on each of the five machines with the fewest number of plays, they would’ve qualified.
GOT: 171 plays, #34 = 54 points
Avengers: 220 plays, #44 = 44 points
WW: 246 plays, #49 = 39 points
DH: 248 plays, #50 = 38 points
TOM: 257 plays, #51 = 37 points
54 + 44 + 39 + 38 + 37 = 212 points
Seems pretty reasonable: you can knock a couple games out of the park, or be very consistent at the required percentile, either way works.
Any chance of seeing the data with the bonus points for 1st and 2nd scaling as well?
So 200, 180, 170, 169, 168…
I understand the thinking behind this, but still trying to form an opinion. My biggest hang-up is:
If PAPA ticket-style qualifying is meant to compare competitors’ entire tickets, then why would we intentionally neglect to compare the relative strength of a significant part (in many cases, seems to be 2/5) of their tickets to the entirety of the competition?
Interesting stuff.
Let’s get to the ‘no spin zone’ here Joe
You cherry picking the 5 games with the lowest number of plays doesn’t paint the most accurate picture IMO.
If you take all 10 games, and do the same math, it’s an average point total of 36.4 per game. This ends up at 182 for 5 games, and puts you in 36th place (top 30% finish in A division, not top 20%). The numbers obviously get worse if you take just the ‘other 5’ games you didn’t use, but I don’t think cherry picking the most played games is a fair evaluation either.
I’m more curious about the difference of this over time, so I pulled PAPA 17 and 18 numbers (first years with 24 qualifiers) . . .
For PAPA 17, taking all 10 games that year, a top 20% score yielded 41.9 points per game. This ends up with a run of 210 points for 5 games, which would put a player in 21st place (top 22% of the field).
For PAPA 18, a top 20% score yielded 41.1 points per game, 206 point total run. This would put a player in 19th place (top 19.8% of the field).
I actually agree with you in theory that it’s completely reasonable to assume if a player does a run where their average finish is in that top 20% of every game, that they SHOULD be finishing in the top 20% of the field.
It looks like that’s normally the case prior to PAPA 19. The question is was this just a crazy IFPAPA year in terms of player count and entries thrown into the mix, or will the numbers continue to grow (in which case finishing in that top 20% of entries on average won’t get you in that top 20% of an overall finish).
I’m newer to all of this, so I’m just thinking out loud here.
The current system puts a premium on top scores by putting a point gap in: 100-95. In an alternate universe where consistency was put at a premium the scoring might be 100-99 but a ticket where all 5 games scored points got a 5 point bonus.
The philosophy of the current scoring system definitely puts a premium on top scores but as a whole that may not have been the case in the past. When player numbers are lower and most reasonable games count for points then adding a premium for top scores creates a balance between top scores and consistency. As more players enter that balance point is moving; in this case towards tickets with top scores but less consistency.
It really comes down to what people think is a better demonstration of skill for qualifying since a scoring system can be made to reflect that philosophy.
Numbers and stuff!
Cool discussion, all.
I just needed to play better!
Added.
Added this, kind of. Started doubling at 49 points.
Belsito Belsito’d himself
Thanks for building these. I kind of really like this double-tail system. That could serve Classics pretty well, too, lessening the horror pain of a one-bad-game ticket. (Cue another swear post from Josh.)
A swear post from Josh if it hurts his chances? Ha! And after he Belsitoed me last year in Classics 3. Maybe we should dock people a point for each loud curse they utter while qualifying. Kevin says we’re not supposed to, right?
I did indeed cherry-pick those games, but I did it because “play the least-played games” is a rational approach that a player can take if they don’t feel they can pop off a top-2 score on any machine. It’s not a secret that it’s easier to get points on a machine that has had fewer games played. (The extreme case would be having Deathball 2000 in the bank, but no one at all is playing it because of that whole “death” problem. But if I’m immortal, I’m totally picking it on my ticket, because that’s free points for me.) Since the play counts are always available to players, and players can choose any games they want on a ticket, this is a viable strategy. If players were assigned machines randomly on each ticket, then sure, an average of all machines in the bank would be more fair.
If people really hate fractions, I’d rather double everything and then use the space to keep the minimum points/place at one. It seems weird to me to have a non-monotonic system. Basically you’re emulating that exponential mentioned earlier, of course.
(I’d just use fractions)
Perhaps it’s just me, but I’ve never had that approach to a PAPA A run . . . ever.
As you’re qualifying Thursday and Friday, how would you even have a sense of which games will ultimately be the least-played games? Or is this an end of day Saturday strategy only?
I can certainly understand it if one of the games has 70 plays on it and you’re guaranteed ‘free points’ (I think back in the day this used to be possible in Classics), but for A division the difference between playing a game with 171 scores on it versus a game with 257 scores on it . . . I just don’t see that as a strategy for success. Getting a 1.8bil on GOT for a top 20% score doesn’t seem any easier compared to putting up 393mil on Tommy. Certainly both are challenging enough for me to want to play the game I feel most comfortable on. But YMMV