Recent Facebook banter regarding fargo

[Jerry R]

Admittedly I mostly skimmed this thread so sorry if I get off base, but I'll just make a few comments.

1. You really can't handicap pool. Maybe in a long race you can get close, but it's silly to think you can handicap something that doesn't have equal opportunities for all players.
Imagine bowling where every time your opponent throws a strike or gets a spare before you, you get a 0 for your score in that frame.
In golf no matter what the other guy does, you still get to hit your ball.

2. If you try to handicap pool and actually went with an aggressive enough "game spot" to even up a match between a 450 and a 700 Fargo, the 700 would not play in the tournament, and oddly enough, it'd be for the same reason the 450 won't play in an open event. They wouldn't think they'd have a chance. Fargo says with a 450 going to three games and a 700 going to 13 games the 700 is still 60% odds to win. How many of these "non-pro" 700 Fargo's would sign up for that event? (Would we tell them to "take their skirt off"?)

3. Capped events tend to work better than handicapped because as noted earlier by many people, handicaps don't really work. The guy who needs 2 or 3 innings to get out will rarely beat the guy who needs 1 or 2 innings no matter how many games they get. But if you cap it so everyone in your bracket needs 2 or 3 innings, then everyone feels they have a chance.

I think the closest anyone came to a true handicap for pool was that very short-lived thing Sigel was involved with where they gave lower skill players ball in hand in the middle of their turn. The theory was that with that system "anyone can run a table". But that still didn't solve the lack of opportunity for the player sitting in their chair. It seemed like it was going pretty well, but I believe the Covid shut down killed it, but I could be off on that timeline.
The other solution I thought was interesting was that "placement pool" deal that Earl used to talk about. Take the randomness of the break away and give every player the same layouts using a projector and score them on how many shots it takes to pocket the balls. Possibly with the lower cost of electronics this could still happen. I'm not saying it's the answer, but it is one of those things I would like to see play out. At lease with the "equal opportunity" format you could develop a golf like handicap and run it as an open event.

 

[nine_ball6970]

I follow a lot of the split bracket tourneys and the 760+ players tend to avoid them minus Rob Saez. Roland IIRC won one of the JOB's tourneys but Bergman played last month and I don't think he won (both go to 13, a 600 fargo would go to like 5, 630 6, etc). These tourneys tend to be won by the low side as much or more than the high side which would go against common sense especially when there could be over 6 figures to play for that should aid the pro not the other way around!

The calcuttas can last for 2 - 3 hours alone and the tourney will be all day sat and LATE into sun which the length you'd assume would benefit the better more seasoned players but you'd be surprised! Indy has several smaller 100ish player split bracket tourneys and Beckley got 3rd or 4th in the last one and he ran a 5 or 6 pack on a Diamond bar table right before the final 4! A guy from the low side manipulated his Fargo because he had 2 names in the system like a 515 and a 540 and he split nearly $6000 for 1st/2nd.

Lots of the lower players are underrated and they dont use fargos for calculating ratings normally. This favors the lower players which leads to huge calcuttas. Known pros are never going to have a rating which is lower than their actual skill. They have to bring their A game every match and that still may not be good enough to win.

 

[BasementDweller]

My only remaining FR question is about player variability. My intuition tells me that the better the player, the closer they'll be to playing at their rating. I'm thinking the lesser player will play over a much wider range. Here I guess, we would have to couple FR and TPA. My problem or question is -- when a 700 plays a 550 and the handicap is close to 50/50, if the 550 plays like a 650 the 700 has zero chance to win. This scenario seems more likely than a 700 magically playing like an 800. Now I'm not talking about their performance rating, which can be extrapolated after the fact by looking at the scores. I'm talking about what actually happens on the table during the match.

When the lesser player plays over their rating, the better player can be justified in feeling like they never had a chance. When this happens enough times, the better players give up. This happened all the time in 9 ball tourneys that used ball handicaps and I'm not sure it won't eventually happen to FR tourneys.

Just sort of thinking out loud here about player variability...

 

[maha]

you have to look at the past tournament winners and how often or well they do compared to lower ranked ones.

if lower ranked players in handicapped tournaments are not close in the results then the system is not fair. and anything that isn't fair cannot last over the long run.

some mentioned the other sports. and that is okay, as in most all the other sports, basically the same players or teams don't win or come in the top few every time like pool or chess. so neither has a following from the public.
if they did as those two the sport would die. as your favorites would always fail and so you would lose interest.

 

[maha]

it is easy to handicap pool it doesn't happen though as the good players want to win all the time and the sport doesn't want to upset them.
it happens every day in the pool room.

weaker players need lots more games on the wire to even make it a contest..

or better yet use a system that sees how often they can run to what ball. and compare that to how often the better player often runs to the nine,
and give that ball as the spot. or something similar that works out better. according to each players chances.
basically it gives every weaker player a real chance to run out to win even if its to the three or four ball.

you could have it where both players win on the nine and one may win on the four ball and one one the say seven. might even make the game more fun to watch. although more complicated.

 

[mikepage]

you have to look at the past tournament winners and how often or well they do compared to lower ranked ones.

if lower ranked players in handicapped tournaments are not close in the results then the system is not fair. and anything that isn't fair cannot last over the long run.

We've done those types of comparisons for long running handicapped tournaments, and the results are as we expect them to be.

Here are the 160 entrants in the "Fairmatch" singles event at BCAPL in Vegas this year. Average rating 492 with a large spread. Matches depend on rating difference and are one of 5-5, 5-4, 6-4, 6-3, 7-3, and 8-3.

I've highlighted the top 32 finishers (all in the $$) and indicated the finish place. These come about one third from players with below average rating and two thirds from players with above average rating.

At first blush, that looks perhaps a little more lopsided toward stronger players than you might expect for "fair" matches where fair means close to 50%/50% expected.

But there are two forces of note at work.

(1) Matchups like 5-5, 5-4, 6-4, 6-3, 7-3 (0, 1, 2, 3, and 4 game spots) don't smoothly blend into one another. You can't give a 2.3 game spot; it's either 2 or 3. So in your effort to make the match close to 50%/50% somebody is going to get an edge when you are forced to lock into one of these matchups. We choose to ALWAYS assign any residual edge to the higher-rated player. That means even these ostensibly 50%/50% matches are in actuality more like on average 55%/45% in favor of the higher rated player. That doesn't sound like such a big difference. But if you look over multiple matches it is bigger. A 55% favorite has twice the chance of winning the next 3-4 matches as does a 45% person

(2) The most lopsided match allowed here, 8-3, is not nearly lopsided enough to get close to 50%/50% when the rating difference is large. A 330-rated player, for instance is at 25% or less chance of winning a match against anybody in the top half of the field.