GLI is No Good

[jsp]

Is it just me, or do you think this GLI (Game Loss Index) tie-breaker business is NOT the best way to go. For GLI, you ONLY look at your losses. Your number of wins is totally irrelevant. That means, for the matches that you lose, it doesn't matter if you win 7 games or 0 games...it's all the same. Let's give an example...

Wins - Losses

Player A
8 - 6
8 - 7
8 - 7
0 - 8
0 - 8
-----------------
24 - 36 (Total)

Player B
8 - 7
8 - 7
8 - 7
7 - 8
7 - 8
-----------------
38 - 37 (Total)

In this situation Player A and Player B both have identical 3-2 records. Player A gets totally dominated in his 4th and 5th match, losing 0 to 8...while Player B goes to hill-hill on each of his matches. From this data, it's reasonable to think that Player B is the better player of the two, because he has a much better win/loss percentage. But by using the GLI index, Player A gets through to the next round because he has one less loss, even though he as 14 less wins. Unfair!

I understand the intent of the IPT by using the GLI as a tie-breaker...to prevent players from purposely losing matches so so-and-so can advance to the next round. However, I'm much more worried about the situation above than instances of cheating.

This is the logic behind GLI... Let's make the system more unfair so that we can minimize cheating. Doesn't that say something about what the organizers think of the integrity of their own pool players? It's kinda ridiculous...especially to the sports fans who are unfamiliar with pool. KT and the organizers want to change pool's image to the general public...but what message does their intenet behind GLI portray? That it's more important to keep the players from cheating than to keep the system totally fair. (And besides, GLI isn't fool-proof against cheating. MikeJanis has already encountered a way to abuse the system.)

Why not just assume the pool players have integrity and would not cheat, and keep the system as fair as possible? I agree you should be penalized by your losses, but you should also be rewarded by your wins. It only makes sense. The best way to go is by using win/loss percentage as the tie-breaker. Not by looking at wins alone and not by looking at losses alone, but by looking at both. Let me know if I'm missing something, because I can't see how GLI would be good for pool.

 

[Gabber]

Is it just me, or do you think this GLI (Game Loss Index) tie-breaker business is NOT the best way to go. For GLI, you ONLY look at your losses. Your number of wins is totally irrelevant. That means, for the matches that you lose, it doesn't matter if you win 7 games or 0 games...it's all the same. Let's give an example...

Wins - Losses

Player A
8 - 6
8 - 7
8 - 7
0 - 8
0 - 8
-----------------
24 - 36 (Total)

Player B
8 - 7
8 - 7
8 - 7
7 - 8
7 - 8
-----------------
38 - 37 (Total)

In this situation Player A and Player B both have identical 3-2 records. Player A gets totally dominated in his 4th and 5th match, losing 0 to 8...while Player B goes to hill-hill on each of his matches. From this data, it's reasonable to think that Player B is the better player of the two, because he has a much better win/loss percentage. But by using the GLI index, Player A gets through to the next round because he has one less loss, even though he as 14 less wins. Unfair!

I understand the intent of the IPT by using the GLI as a tie-breaker...to prevent players from purposely losing matches so so-and-so can advance to the next round. However, I'm much more worried about the situation above than instances of cheating.

This is the logic behind GLI... Let's make the system more unfair so that we can minimize cheating. Doesn't that say something about what the organizers think of the integrity of their own pool players? It's kinda ridiculous...especially to the sports fans who are unfamiliar with pool. KT and the organizers want to change pool's image to the general public...but what message does their intenet behind GLI portray? That it's more important to keep the players from cheating than to keep the system totally fair. (And besides, GLI isn't fool-proof against cheating. MikeJanis has already encountered a way to abuse the system.)

Why not just assume the pool players have integrity and would not cheat, and keep the system as fair as possible? I agree you should be penalized by your losses, but you should also be rewarded by your wins. It only makes sense. The best way to go is by using win/loss percentage as the tie-breaker. Not by looking at wins alone and not by looking at losses alone, but by looking at both. Let me know if I'm missing something, because I can't see how GLI would be good for pool.

Not only that but,
"Corr -- and all the players who made it to day two -- will take their Games Lost Index with her into competition today, and the GLI stat becomes more important as the field is narrowed to the best of the best players."


Gabber ?

 

[Rude Dog]

Good point JSP. GLI needs some work, that's for sure.

 

[jsp]

Not only that but,
"Corr -- and all the players who made it to day two -- will take their Games Lost Index with her into competition today, and the GLI stat becomes more important as the field is narrowed to the best of the best players."

Exactly! A point that I forgot to bring up. Carrying your GLI with you to the next round makes it even more unfair to the players who came from a very difficult first round group (such as Manalo's group). Again, this move makes the system more unfair for the purpose of minimize cheating.

 

[Jaden]

I think GLI is a great thing.

GLI in a round robin format allows for something that isn't available in other tournament formats. It allows for how well you play to be a deciding factor and to be taken into account.

Because everyone plays everyone within their bracket, they have to play eqaully well against the good players that they beat as the other players who beat that same player. If allison beats Feijin 8-6 but archer beats feijin 8-2, then archer gets the better rating. That's the way it should be. It;s not just about beating them it's about how well you played when you beat them and it's something that most pool tournaments have been lacking.

Plus there's the added benefit of controversy which allows for more exciting times and better viewership which can only be good for the sport.

I personally think they should also include how many games you won in the matches you lost. That would make it even more accurate.

If they had GWI as well as GLI and averaged the two for the final decisions, I think it would be awesome.

 

[jsp]

If allison beats Feijin 8-6 but archer beats feijin 8-2, then archer gets the better rating. That's the way it should be. It;s not just about beating them it's about how well you played when you beat them and it's something that most pool tournaments have been lacking.

I agree, but win/loss percentage would answer this question as well. My question is why not just use win/loss percentage instead of just looking at losses?

 

[mnShooter]

Jaden you are exactly right. There should be a Power Ranking calculated as

PR = GWI - GLI

The higher ranking the better.

So wins and losses are calculated. This would be much more accurate and fair IMO. Not sure how it would affect the taking a dive aspect of cheating though.

 

[onepocketchump]

Is it just me, or do you think this GLI (Game Loss Index) tie-breaker business is NOT the best way to go. For GLI, you ONLY look at your losses. Your number of wins is totally irrelevant. That means, for the matches that you lose, it doesn't matter if you win 7 games or 0 games...it's all the same. Let's give an example...

Wins - Losses

Player A
8 - 6
8 - 7
8 - 7
0 - 8
0 - 8
-----------------
24 - 36 (Total)

Player B
8 - 7
8 - 7
8 - 7
7 - 8
7 - 8
-----------------
38 - 37 (Total)

In this situation Player A and Player B both have identical 3-2 records. Player A gets totally dominated in his 4th and 5th match, losing 0 to 8...while Player B goes to hill-hill on each of his matches. From this data, it's reasonable to think that Player B is the better player of the two, because he has a much better win/loss percentage. But by using the GLI index, Player A gets through to the next round because he has one less loss, even though he as 14 less wins. Unfair!

I understand the intent of the IPT by using the GLI as a tie-breaker...to prevent players from purposely losing matches so so-and-so can advance to the next round. However, I'm much more worried about the situation above than instances of cheating.

This is the logic behind GLI... Let's make the system more unfair so that we can minimize cheating. Doesn't that say something about what the organizers think of the integrity of their own pool players? It's kinda ridiculous...especially to the sports fans who are unfamiliar with pool. KT and the organizers want to change pool's image to the general public...but what message does their intenet behind GLI portray? That it's more important to keep the players from cheating than to keep the system totally fair. (And besides, GLI isn't fool-proof against cheating. MikeJanis has already encountered a way to abuse the system.)

Why not just assume the pool players have integrity and would not cheat, and keep the system as fair as possible? I agree you should be penalized by your losses, but you should also be rewarded by your wins. It only makes sense. The best way to go is by using win/loss percentage as the tie-breaker. Not by looking at wins alone and not by looking at losses alone, but by looking at both. Let me know if I'm missing something, because I can't see how GLI would be good for pool.

I agree, the W/L % shoudl be the tie breaker. It would also accomplish the same thing as more wins and less losses would be the reason not to cheat.

John

 

[Snapshot9]

Why ...

Is it just me, or do you think this GLI (Game Loss Index) tie-breaker business is NOT the best way to go. For GLI, you ONLY look at your losses. Your number of wins is totally irrelevant. That means, for the matches that you lose, it doesn't matter if you win 7 games or 0 games...it's all the same. Let's give an example...

Wins - Losses

Player A
8 - 6
8 - 7
8 - 7
0 - 8
0 - 8
-----------------
24 - 36 (Total)

Player B
8 - 7
8 - 7
8 - 7
7 - 8
7 - 8
-----------------
38 - 37 (Total)

In this situation Player A and Player B both have identical 3-2 records. Player A gets totally dominated in his 4th and 5th match, losing 0 to 8...while Player B goes to hill-hill on each of his matches. From this data, it's reasonable to think that Player B is the better player of the two, because he has a much better win/loss percentage. But by using the GLI index, Player A gets through to the next round because he has one less loss, even though he as 14 less wins. Unfair!

I understand the intent of the IPT by using the GLI as a tie-breaker...to prevent players from purposely losing matches so so-and-so can advance to the next round. However, I'm much more worried about the situation above than instances of cheating.

This is the logic behind GLI... Let's make the system more unfair so that we can minimize cheating. Doesn't that say something about what the organizers think of the integrity of their own pool players? It's kinda ridiculous...especially to the sports fans who are unfamiliar with pool. KT and the organizers want to change pool's image to the general public...but what message does their intenet behind GLI portray? That it's more important to keep the players from cheating than to keep the system totally fair. (And besides, GLI isn't fool-proof against cheating. MikeJanis has already encountered a way to abuse the system.)

Why not just assume the pool players have integrity and would not cheat, and keep the system as fair as possible? I agree you should be penalized by your losses, but you should also be rewarded by your wins. It only makes sense. The best way to go is by using win/loss percentage as the tie-breaker. Not by looking at wins alone and not by looking at losses alone, but by looking at both. Let me know if I'm missing something, because I can't see how GLI would be good for pool.

Don't they just use
games won / games played = winning percentage (xx.xxx%)

so in the example above player A won 24 of 60 games = 40% win percentage
player B won 38 of 75 games = 50% win percentage

Therefore Player B advances and Player A does not. (A is rightfully punished
for getting zipped 2 matches, and Player B rewarded for losing 2 hill-to-hill
matches given that each won 3 of their matches).

This winning percentage would, in effect, become their 'batting average'.

 

[breakup]

I see you guys beat me to the punch, I was working on this image and you all got it posted first , anyway here is an interesting twist to what has already been posted.
Under this format winning the 15 game in a hill / hill match is no more or less important than the first game. There is really no premium on winning the match, just one more game. As you can see in the hypothetical round robin with 6 of the 7 dwarfs (Sneezy has conflicting contractual obligations) example (if I did my math correctly) Bashful could win all his matches and not only fail to advance but actually have the worst score in the group!!

 

[Snapshot9]

I like your style ...

I see you guys beat me to the punch, I was working on this image and you all got it posted first , anyway here is an interesting twist to what has already been posted.
Under this format winning the 15 game in a hill / hill match is no more or less important than the first game. There is really no premium on winning the match, just one more game. As you can see in the hypothetical round robin with 6 of the 7 dwarfs (Sneezy has conflicting contractual obligations) example (if I did my math correctly) Bashful could win all his matches and not only fail to advance but actually have the worst score in the group!!

LMAO ... Do the top 3 advance to Snow White? .... lol

 

[mnShooter]

Don't they just use
games won / games played = winning percentage (xx.xxx%)

Yeah this would be much simpler than my calculation with the same result.

Breakup- bashful would still win because the number of wins is the first deciding factor - GLI is second.

 

[NaClBandit]

I was under the impression that the first criterium is matches won, followed by GLI to break ties. I could be wrong though.

I think if 2 players are tied after GLI then the first one to execute Massey's "Boot Shot" advances.

 

[TheBook]

I love Snow White. Didn't Disney really pull a fast one over the censors with this feature. Every morning the dwarfs greet Snow White by singing "Hi Ho!"

 

[breakup]

Breakup- bashful would still win because the number of wins is the first deciding factor - GLI is second.

Thanks!! I was hoping I was overlooking something.

 

[Snapshot9]

Yeah ....

I love Snow White. Didn't Disney really pull a fast one over the censors with this feature. Every morning the dwarfs greet Snow White by singing "Hi Ho!"

ROFL ... but I wouldn't have called Snow White work, been more
like Pleasure ... lol (now quit thinking of those nasty cartoons they have)

 

[Gabber]

Its a system that invites abuse.
Normally, if a player throws a match, it costs him because he loses the match.
With this system, a player can 'throw' the match and still win the game! All he has to do is allow his opponent enough games so that they both go thru to the next round!
Either they havent thought this thru or they have!

Gabber

 

[jsp]

the best way to prevent tie-breaking cheating

If you want a full-proof way to eliminate tie-breaking cheating in a round roubin format, the number of break and runouts should be used as the first tie breaker (instead of GLI, GWI, winning percentage...etc) and the format should be alternating break. Therefore, it's impossible to conspire the outcome of a tie-breaker. You have to put in the work and run the racks yourself to come out on top of the tie-breaker. The opponent would have absolutely no power to decide that number. You can have Efren Reyes or Bugs Bunny as your opponent, it doesn't matter, it's all on you to break and runout.

It's kind of interesting to think about what it would be like if they adopt a system like this. Players who are on the bubble would be much more aggressive in terms of trying to runout the rack. They would go for difficult breakout shots or attempt impossible bank or kick shots just to try to sustain a run.

Actually, come to think about it, it wouldn't be a perfect system. If someone is up 5-0 and needs 3 more break and runouts to advance to the next round (plus win the match), he'd probably pot the 8 ball prematurely on purpose for a couple games just so that he can obtain more opportunities to break and runout. Technically, that wouldn't be cheating...but it would be weird to see. If his opponent really didn't want him advancing, then he would probably pot the 8 himself and lose a rack on purpose so his opponent won't have anymore chances. Hehe, it's not perfect, but at least there would be no conspiracies.

 

[jay helfert]

Tie Breaker

Is it just me, or do you think this GLI (Game Loss Index) tie-breaker business is NOT the best way to go. For GLI, you ONLY look at your losses. Your number of wins is totally irrelevant. That means, for the matches that you lose, it doesn't matter if you win 7 games or 0 games...it's all the same. Let's give an example...

Wins - Losses

Player A
8 - 6
8 - 7
8 - 7
0 - 8
0 - 8
-----------------
24 - 36 (Total)

Player B
8 - 7
8 - 7
8 - 7
7 - 8
7 - 8
-----------------
38 - 37 (Total)

In this situation Player A and Player B both have identical 3-2 records. Player A gets totally dominated in his 4th and 5th match, losing 0 to 8...while Player B goes to hill-hill on each of his matches. From this data, it's reasonable to think that Player B is the better player of the two, because he has a much better win/loss percentage. But by using the GLI index, Player A gets through to the next round because he has one less loss, even though he as 14 less wins. Unfair!

I understand the intent of the IPT by using the GLI as a tie-breaker...to prevent players from purposely losing matches so so-and-so can advance to the next round. However, I'm much more worried about the situation above than instances of cheating.

This is the logic behind GLI... Let's make the system more unfair so that we can minimize cheating. Doesn't that say something about what the organizers think of the integrity of their own pool players? It's kinda ridiculous...especially to the sports fans who are unfamiliar with pool. KT and the organizers want to change pool's image to the general public...but what message does their intenet behind GLI portray? That it's more important to keep the players from cheating than to keep the system totally fair. (And besides, GLI isn't fool-proof against cheating. MikeJanis has already encountered a way to abuse the system.)

Why not just assume the pool players have integrity and would not cheat, and keep the system as fair as possible? I agree you should be penalized by your losses, but you should also be rewarded by your wins. It only makes sense. The best way to go is by using win/loss percentage as the tie-breaker. Not by looking at wins alone and not by looking at losses alone, but by looking at both. Let me know if I'm missing something, because I can't see how GLI would be good for pool.

This is an excellent observation! A simple way to break ties (if two players have identical W/L records) would be to just add up the total number of games won.

 

[SlimShafty]

I agree some hybrid system may be more accurate of a players ability, but the GLI is easy for the average viewer to follow, but like all of you I still am very curious why a plain old Winning % is not used?

But so far even if you used a winning % (like all major sports) with all the players that were tied, the outcome for at least this round would have been the same.

Hundal, Raj*_____GLI 5.00 Winning % .545
Feijen, Niels*____GLI 5.20 Winning % .574
Fisher, Allison____GLI 5.80 Winning % .517

Hohmann, Thorsten*__5.00 Winning % .569
Robles, Tony*________5.20 Winning % .581
Jones, Jeremy________6.60 Winning % .500

Williams, Charlie*_____6.80 Winning % .477
Souquet, Ralf________7.00 Winning % .462
San, Souci George____7.20 Winning % .455
McCready, Keith______7.40 Winning % .439

Of course I doubt this will hold up over the entire tournament, but I still think this system creates some exciting match-ups, (even though I would have opted for the winning %) if your palying a guy with a low GLI and your close to making it and your GLI is high, you better knuckle and win.