5 SVB US open titles - 5 points

[pt109]

Apple to Oranges.

Not going for the old 'apples and oranges'.

The two most exciting matches in thei Open were the ones Shaw made comebacks in.
....sheer skill...

These were not going to happen at alternate break because the comebacks would require
the opponent's cooperation.

Sometimes, no matter how good you are, you gotta sit in 'the martry's chair'....
...which makes you cherish your turn at the table all the more.

Take a walk on the wild side, Don....it's exhilarating.

 

[Poolshootindon]

Not going for the old 'apples and oranges'.

The two most exciting matches in thei Open were the ones Shaw made comebacks in.
....sheer skill...

These were not going to happen at alternate break because the comebacks would require
the opponent's cooperation.

Sometimes, no matter how good you are, you gotta sit in 'the martry's chair'....
...which makes you cherish your turn at the table all the more.

Take a walk on the wild side, Don....it's exhilarating.

On the other side. If Shaw had not played poorly he would not have been so far behind.

So lets see if this makes sense. A NFL team plays poorly for 3/4's of the game an they are down 35 to 0. They finally score a field goal. Should they be given the ball again so they can make a comeback?

 

[CreeDo]

I guess I was not clear in what I said above. According to an analysis of the probabilities involved, the better player will have exactly, and I mean exactly, the same chance of winning against any level of competition with either format. Let me state the same thing a different way. If Shane is a 74.37% expected winner against Joe under a winner-breaks format, his is a 74.37% expected winner against Joe under an alternate-breaks format. I did not expect this to be the case when I looked at it, but that's how the probabilities work out.

This analysis of course ignores psychological issues like the fabled "momentum" phenomenon in sports.

I had someone in a [math forum](https://m.reddit.com/r/statistics/comments/44b99h/trying_to_calculate_whether_winner_breaks_vs/) do an analysis and they concluded that winner breaks very slightly favors the stronger player (assuming there is a stronger player).

Not being a math wiz, I can't comment on whether their work is accurate or not.

It makes sense to me though.

Basically, any rack that isn't a break and run could be thought of as a coin flip. And in alternate breaks, if both players are equal at running racks, it's still a coin flip... just with a similar number of runouts added to both sides.

But in a winner breaks format, the winner gets a shot at performing an extra break and run that the other player is denied. If he fails, his odds then revert to 50/50 again. So before that typical 50/50 outcome, he has a 'free roll' at an uncontested win.

Does that have any holes in the logic?

 

[CreeDo]

I think there's a format where alt. breaks works well and can even be more exciting than winner breaks, and that's in games where the player is expected to win 80, 90% of the time from the break.

Then, instead of having that excitement of "will he get out here?" (which is only 20% likely) the excitement comes from "will he drop the ball and FAIL to get out here? (which again is only about 20% likely, or less).

That makes sense for 8 ball, with open after the break rules and a consistent rack. Might also make sense for 7-ball. I really enjoyed us open 8b under those rules.

 

[Matt]

I had someone in a [math forum](https://m.reddit.com/r/statistics/comments/44b99h/trying_to_calculate_whether_winner_breaks_vs/) do an analysis and they concluded that winner breaks very slightly favors the stronger player (assuming there is a stronger player).

Not being a math wiz, I can't comment on whether their work is accurate or not.

It makes sense to me though.

Basically, any rack that isn't a break and run could be thought of as a coin flip. And in alternate breaks, if both players are equal at running racks, it's still a coin flip... just with a similar number of runouts added to both sides.

But in a winner breaks format, the winner gets a shot at performing an extra break and run that the other player is denied. If he fails, his odds then revert to 50/50 again. So before that typical 50/50 outcome, he has a 'free roll' at an uncontested win.

Does that have any holes in the logic?

Is that "extra" break any different than the extra break that the winner of the lag gets in an alternate break set that ends with an odd number of games played?

Because of the extra break opportunity, it would be more fair if you had to win by two, like in most games that have a serve involved, but then you run the risk of a close match going on indefinitely.

 

[justaplayer]

I guess I was not clear in what I said above. According to an analysis of the probabilities involved, the better player will have exactly, and I mean exactly, the same chance of winning against any level of competition with either format. Let me state the same thing a different way. If Shane is a 74.37% expected winner against Joe under a winner-breaks format, his is a 74.37% expected winner against Joe under an alternate-breaks format. I did not expect this to be the case when I looked at it, but that's how the probabilities work out.

This analysis of course ignores psychological issues like the fabled "momentum" phenomenon in sports.

From my experience alternating break formats lead to much closer score, many more hill hill matches...and therefore more upsets.

As you said just the deflating effect of a guy running out 5 racks has to affect outcomes, beyond the score. Sometimes when you are playing strong it seems there is only room enough for one guy to play at the highest level.

 

[BRussell]

I think there's a format where alt. breaks works well and can even be more exciting than winner breaks, and that's in games where the player is expected to win 80, 90% of the time from the break.

Then, instead of having that excitement of "will he get out here?" (which is only 20% likely) the excitement comes from "will he drop the ball and FAIL to get out here? (which again is only about 20% likely, or less).

That makes sense for 8 ball, with open after the break rules and a consistent rack. Might also make sense for 7-ball. I really enjoyed us open 8b under those rules.

I agree it's got to be alternate break on a small table with minimal break requirements like the US Bar Table and US Open (bar table) 8-ball, and it becomes like tennis to see who's going to break serve.

But I'd much prefer these US Open 9-ball rules with tougher break requirements and winner breaks. The games have more variety, and there are good possibilities for packages and come-backs.

 

[Dockter]

One thing I'm not sure has been mentioned is the length of the race.

Winner breaks would be better in a longer race (9+). Reason being, it gives a trailing player a better chance to maintain control when mounting a comeback, and the likelihood of running out the set from the lag is extremely rare.

When you have shorter races (7 or under), then I think you have to play alt break as running out the set becomes a bigger possibility.

This. Does anyone really want to see winner breaks in a race to 5 or 7? The set is easily ran out by a number of these guys. If someone runs out it's great shooting but not super impressive for a pro especially with a 9 on the snap and a couple early combos.

Now if you run out 9 or 11, that's impressive and you deserve the win. I love winner breaks because of the comebacks. The pressure of being up there and knowing this could be your last shot at the table if you don't play a good shot has to be intense and that comeback by Shaw was one of the greats.

 

[CreeDo]

Is that "extra" break any different than the extra break that the winner of the lag gets in an alternate break set that ends with an odd number of games played?

Because of the extra break opportunity, it would be more fair if you had to win by two, like in most games that have a serve involved, but then you run the risk of a close match going on indefinitely.

Yeah the lag was brought up I think in the reddit thread as possibly having the biggest influence on the overall winner. Would be cool if accu-stats had kept stats about lags this entire time, and we could see how often the winner of the lag won the set.

The extra break you get from winner's break is different from the lag in that it can happen an indefinite number of times, i.e. you can string consecutive racks (which we often see... someone screws up and gets punished not just in the current game, but one or more additional games).

In any case, longer races wipe out most of the advantage from lag or winner break format.

 

[DelawareDogs]

I also like alt break cos it is fairer.
Winners break is only exciting if there are packages which is rare if the breaks rules and tables breaks tough like this US Open
whereas alt break is more intense as each player focuses on holding his break and not letting other player steals his break. In winners break, player can get complacent thinking he can catch up no matter how far behind and the chance to comeback may never come. Incredible comebacks like Shaw v Ko Pin Yi and Chang are very rare :smile:

No way!!! I absolutely disagree. Packages are rare in that even the players who can string packages don't always get the opportunity.

But this doesn't make alt break any more intense. You're right in the point that in winner break, a player can get complacent and think they can catch up, digging themselves further into a hole. The complacent player shouldn't be the one winning anyway....:thumbup:

In alternate break each player focuses on holding his break.....
That's all fine and dandy - In winner break each player tries to hold on as long as they can, because in the heat of tournament like that, they know there's a dynamite player waiting in that chair that can put a clinic on. Ready to pounce as soon as they miss....

It's disconcerting and off-putting to know that after I step to the table, break and run a nice rack of nine ball, ten ball, eight ball...... my reward is to take my seat and watch my opponent attempt the same thing???
Why? What did he do to earn that?

I say let 'em shoot. When they miss then let the other person shoot. Alternate break is only fairer if you are the person that can only break and run one rack, or are the person who needs to break to feel like he's part of the game instead of sitting in a chair for 35 minutes. It's not fair to the person who can string a pack, but instead is forced to take a seat once he does his job.

 

[BeiberLvr]

No way!!! I absolutely disagree. Packages are rare in that even the players who can string packages don't always get the opportunity.

But this doesn't make alt break any more intense. You're right in the point that in winner break, a player can get complacent and think they can catch up, digging themselves further into a hole. The complacent player shouldn't be the one winning anyway....:thumbup:

In alternate break each player focuses on holding his break.....
That's all fine and dandy - In winner break each player tries to hold on as long as they can, because in the heat of tournament like that, they know there's a dynamite player waiting in that chair that can put a clinic on. Ready to pounce as soon as they miss....

It's disconcerting and off-putting to know that after I step to the table, break and run a nice rack of nine ball, ten ball, eight ball...... my reward is to take my seat and watch my opponent attempt the same thing???
Why? What did he do to earn that?

I say let 'em shoot. When they miss then let the other person shoot. Alternate break is only fairer if you are the person that can only break and run one rack, or are the person who needs to break to feel like he's part of the game instead of sitting in a chair for 35 minutes. It's not fair to the person who can string a pack, but instead is forced to take a seat once he does his job.

Packages are rare.

But winner breaks is more pressure because if you miss the other guy might put up a package.

But packages are rare.

 

[Bob Jewett]

I'm sure that's true using a binomial analysis, but that assume independence. Did you look at this taking into account the non-independence of the events? Or did you use an empirical analysis?

It was based on each player having a certain percentage win from the break with those adding up to more than 100%. That's different from the independent assumption.

 

[Bob Jewett]

For those who are interested in the calculations for alternate breaks versus winner breaks, I made an Excel spreadsheet that lets you calculate the odds in each case.

The inputs are the "wins-from-the-break" percentages of the two players when playing each other. For example, SVB might be 65% to win on his breaks and Dennis Orcullo might be 55% when they play each other. The usual (and too simple for this purpose) assumption usually used is that who breaks makes no difference. The reasons that assumption is used are first that it is much easier to calculate the match odds.

The second is that it takes a lot more work to find the statistics for wins-from-the-break percentages. You can't simply look at the match score; you have to track each game. AtLarge has done this for a lot of matches.

One thing that a lot of people overlook is that the break is only a small advantage if any to most pro players. Pat Fleming did the analysis for many pro players back in the 1980s and in most cases the break was a disadvantage. That is, the players were less than 50% to win a game from their break. In the recently completed US Open, the breaker won the game only 55% of the time (AtLarge's statistics), and that was in a winner breaks format where you expect a strong player to string games.

If anyone wants the spreadsheet to do their own calculations, I'd be glad to send you a copy.

Here is an example of what the sheet looks like for the case of alternate breaks. The numbers in the shaded grids are the chances each score will occur.

 

[Petros Andrikop]

Still the question remains:
Does a player like SVB have equal chances winning a continuous break tournament compared to winning alternate break tournaments? It doesn't appear so...

 

[Bob Jewett]

From my experience alternating break formats lead to much closer score, many more hill hill matches...and therefore more upsets. ... .

It does lead to closer matches if the break is an advantage, but as the statistics have shown, the break is not much of an advantage and is actually a disadvantage for some pro-level players.

But in an extreme case, imagine two players are each 99% to win from the break. In a match to 10 and winner breaks, the player who wins the lag is about 90% favored to win the match because he is about 90% to run out the whole match. If he loses any game, the other player is very likely to run out, and the first shooter will lose one of 10 games about 10% of the time (1% in each game). The match will likely end up 10-0 about 90% of the time.

On the other hand, with alternate breaks, the player who wins the lag will have one extra break, but again on each game where he breaks he has a 1% chance of losing the game and so he might lose his extra break advantage. The match will likely end up 10-9 (about 80% of the time).

 

[Bob Jewett]

Still the question remains:
Does a player like SVB have equal chances winning a continuous break tournament compared to winning alternate break tournaments? It doesn't appear so...

Unless you assume some kind of psychological wrinkle, the odds are the same.

 

[TCo]

Thank you for your reply, always a pleasure to have input from people with deep knowledge of the game.
The stats released after another strong tournament have demolished the myth of "there won't be enough chances for both players in continuous break format", you are right that in the long run better players still win more, however:
- X-packs remain more impressive to the spectators
- On continuous break better players would rarely lose a match they are not supposed to against an inferior player. Better players are not protected anymore and this is not good for the game, not in big tournaments.

As a spectator I disagree. I can track how many have they both run. I think this adds pressure to both players with the alternating. Then a xpack is even more impressive imo.

Appreciate the viewpoint on reacting and overreacting. We seem to be a society that needs to be right always. Polarizing in a grayish world.

 

[efirkey]

The most exciting matches to watch at this year's Open was Shaw coming back from 10 - 4 down not once but almost a second time. This would never happen with alternate break.

 

[Matt]

Bob, thanks for sharing your results. It's always nice to see the numbers back up the logic. If you really want to see that spreadsheet blow up, throw in some handicaps. :smile:

This reminds me of one of the things that I've told league teammates before: Just being favored to win doesn't mean you will win, so you just play your best game to swing the odds your way and don't worry so much about the outcome.

To get back on topic (winner breaks vs. alt. breaks), I'm curious if you prefer one format over the other after having done that analysis. Myself, I think the winner break format is good for a professional level tournament where keeping the audience engaged is a consideration because of the possibility of packages and shorter matches, but the alternate break format is good for a tournament where you want to make sure everyone gets a chance to play and don't have as much of a time constraint.

 

[Petros Andrikop]

Unless you assume some kind of psychological wrinkle, the odds are the same.

I respectfully disagree, results over the last years show something different. He's more likely winning a US open than any other major international event. Breaking format has to be considered here.