Top 57 Players on Earth

[railbird99]

There was NO criticism of Fargo in my post-ZERO!! None!!

My apologies, it seemed like you were saying the fargo system in particular is flawed because won't have all possible data. Plus all the questions seemed to like you were trying to be critical.

I personally don't know anything about fargo except that it's an ELO based system, and I've been a proponent for using ELO to rate pool players forever. It would be awesome to have a unified ELO rating system some day.

 

[iusedtoberich]

The equal break chances discussion...

I think at the top level, lets say Shane and Wu playing winner break 9 ball with magic rack, where they will hypothetically run a 5 pack every time they come to the table... If that were to happen, I think the system would still balance out. In that case, every inning of a player is worth 5 games. But since they are both running 5 every inning, it balances out. They both had equal opportunities.

Now, lets say it was a race to 10. Wu runs 5, then Shane runs 5, then Wu runs 5. Match over, Wu wins 10-5. So this might seem unfair because Wu had 2 innings,
Shane had 1. But over the course of many matches and many tournaments, they will equal out in who ran the first 5 pack, so they would both have equal innings over the long haul.

Now, lets say Wu, in our hypothetical example, plays Corey. Wu runs 5 racks every inning, Corey runs 2 racks every inning. So in our race to 10, Wu runs 5, Corey runs 2,
Wu runs 5. Game over, Wu wins 10-2. They repeat this over and over. On the sets that Corey goes first, it goes Corey runs 2, Wu runs 5, Corey runs 2, Wu runs 5. Game over, Wu wins 10-4. That's still ok, because the better player gets more "games" every time he steps to the table, because he is a higher skill level. The only way for Corey to get the same amount of "games" is to run more racks every time its his inning. If he does not, then he is clearly not as good as Wu, and thus his rating is appropriately lower.

So, I don't think winner break vs alternate break really makes any difference. All that matters is who wins each game.

Bringing this back to the chess Elo for further discussion, if 2 players play a match, do both have equal opportunities as white? Also, lets say two players only play one single game, then go home. One player will obviously have the advantage being white. Does that game count in the chess Elo ratings?

 

[Poolplaya9]

The equal break chances discussion...

I think at the top level, lets say Shane and Wu playing winner break 9 ball with magic rack, where they will hypothetically run a 5 pack every time they come to the table... If that were to happen, I think the system would still balance out. In that case, every inning of a player is worth 5 games. But since they are both running 5 every inning, it balances out. They both had equal opportunities.

Now, lets say it was a race to 10. Wu runs 5, then Shane runs 5, then Wu runs 5. Match over, Wu wins 10-5. So this might seem unfair because Wu had 2 innings,
Shane had 1. But over the course of many matches and many tournaments, they will equal out in who ran the first 5 pack, so they would both have equal innings over the long haul.

Now, lets say Wu, in our hypothetical example, plays Corey. Wu runs 5 racks every inning, Corey runs 2 racks every inning. So in our race to 10, Wu runs 5, Corey runs 2,
Wu runs 5. Game over, Wu wins 10-2. They repeat this over and over. On the sets that Corey goes first, it goes Corey runs 2, Wu runs 5, Corey runs 2, Wu runs 5. Game over, Wu wins 10-4. That's still ok, because the better player gets more "games" every time he steps to the table, because he is a higher skill level. The only way for Corey to get the same amount of "games" is to run more racks every time its his inning. If he does not, then he is clearly not as good as Wu, and thus his rating is appropriately lower.

I think you are right. I also think that over the long haul for typical tournament length matches both breaking formats will show about the same win loss percentages between the two players. Over the short haul though alternate break would be slightly more accurate in determining the best player. But since this system will have a fairly large sample size I don't think it probably makes much difference that the two formats are included without distinction.

So, I don't think winner break vs alternate break really makes any difference. All that matters is who wins each game.

I think you would be right if Fargorate only used who won or lost for their ratings but if I recall correctly they use what the actual score was as compared to what it was expected to be based on the player's rankings (as in I think your rating is impacted differently when you lose to Orcollo 9-8 as opposed to if you lose to him 9-0). Since actual score and not just who won makes a difference, I think in small sample sizes whether they were alternate or winner breaks does make a difference since one creates larger disparities in the score differences than the other. But for most active players the sample size will be big enough that it will likely not be much of a factor because everyone starts getting similar a similar proportion of both kinds and it all kind of evens out.

 

[SUPERSTAR]

Bringing this back to the chess Elo for further discussion, if 2 players play a match, do both have equal opportunities as white? Also, lets say two players only play one single game, then go home. One player will obviously have the advantage being white. Does that game count in the chess Elo ratings?

Most tournaments are round robin, they will typically get 2 games against each opponent. Once as black, once as white. Although some round robins are only partial and don't always result in 2 players playing each other twice.
All head to head matches are set colors for each game, switching to the other color for the next game for a set duration of games till an appropriate winning score is attained.
And yet other tournaments are knockout type events with each pairing mimicking head to head matches, with the winner of each match advancing to the next round and repeating the process.
But for the most part, regardless of the format, everyone gets a chance to play equally with both the white and black pieces.

 

[railbird99]

The equal break chances discussion...

I think at the top level, lets say Shane and Wu playing winner break 9 ball with magic rack, where they will hypothetically run a 5 pack every time they come to the table... If that were to happen, I think the system would still balance out. In that case, every inning of a player is worth 5 games. But since they are both running 5 every inning, it balances out. They both had equal opportunities.

Now, lets say it was a race to 10. Wu runs 5, then Shane runs 5, then Wu runs 5. Match over, Wu wins 10-5. So this might seem unfair because Wu had 2 innings,
Shane had 1. But over the course of many matches and many tournaments, they will equal out in who ran the first 5 pack, so they would both have equal innings over the long haul.

Now, lets say Wu, in our hypothetical example, plays Corey. Wu runs 5 racks every inning, Corey runs 2 racks every inning. So in our race to 10, Wu runs 5, Corey runs 2,
Wu runs 5. Game over, Wu wins 10-2. They repeat this over and over. On the sets that Corey goes first, it goes Corey runs 2, Wu runs 5, Corey runs 2, Wu runs 5. Game over, Wu wins 10-4. That's still ok, because the better player gets more "games" every time he steps to the table, because he is a higher skill level. The only way for Corey to get the same amount of "games" is to run more racks every time its his inning. If he does not, then he is clearly not as good as Wu, and thus his rating is appropriately lower.

So, I don't think winner break vs alternate break really makes any difference. All that matters is who wins each game.

Bringing this back to the chess Elo for further discussion, if 2 players play a match, do both have equal opportunities as white? Also, lets say two players only play one single game, then go home. One player will obviously have the advantage being white. Does that game count in the chess Elo ratings?

I understand where you're coming from, but again, this is not the right mindset. ELO ratings don't measure skill, at least not directly. It measures the probability of a win or loss between two players.

Let's say Shane is the best player in the world. He can play a perfect set and still lose, right? So, because of luck and lack of opportunities, his probability of winning (even when he's playing great) is less than 100% if he's playing someone less skilled than him. Since ELO is measuring probability of winning/losing, his ELO will be less than perfect because of those outside factors.

This is why it does make sense for a person's ELO to be lowered even if they don't get to shoot once in a set. It reflects the probability of this situation happening in the future. This is what we are trying to predict. We aren't asking is Player A better than Player B? We are asking who's going to win between Player A and Player B? The winner is definitely not always the best player.

Over time, with enough data, skill is going to be the main determinant of your ELO rating. Those losing scores (and winning scores) that were because of luck will be insignificant. It will be an accurate measure of how good you are, relative to other players.