The Blind Structure Solution

by , Sep 17, 2007 | 11:52 am

I’m tired of the complaining about blind structures, including my own whiny little self regarding WPT final table structures. Rather than continue to complain, I have decided to propose a new way of looking at tournament structures.

In order to solve the problem, we first have to define it. In my opinion, the problem is that up until now it has been impossible to please everyone regarding structure. I have been in a tournament talking about the structure of the tournament with the players at my table, and one says, “This tournament is too slow, I like a faster structure, so I don’t have to sit around all day.” Another player says, “I like the structure, it gives me a chance to play”, and believe it or not, a guy across the table says, “This is too fast, we are going to lose half the field after 4 levels.”

I have a couple of concerns about our existing structures. Even though the structures are printed before the tournament, players don’t know what they mean in terms of average stack size compared to the blinds at each level. Also, it’s hard to determine how fast players will be eliminated based upon just seeing the structure sheet. The tournament might have been planned for a slow structure and turn into a fast one based upon the timing of player eliminations.

I think it would be great to know that when you are in a tournament, the average stack size as compared to the blinds will most always be a constant. Therefore, before you enter a tournament, you would know that this tournament will adjust the blinds only after enough people have been eliminated in order to justify an increase in the blinds to keep the same ratio of average stack size to big blind.

For example, at the main event at the WSOP, each player starts out with 20,000 in chips and the blinds are 50-100. This means that at the beginning of the tournament, the ratio of average chips to big blind is 200. Therefore, the average stack size has to be 30,000 prior to the blinds moving to 150-300. The increase in blinds is based upon the number of people eliminated, not the clock. Now instead of having a blind structure sheet, tournament directors would say that the main event is a “200-times structure.” Now you know everything you need to know.

If you are playing in a 10-times structure tournament, you know that it is a real fast tournament. The average stack will always be 10 times the big blind. Very few tournaments would be a 200-times structure, and maybe the main event wouldn’t be either. Other alternatives would be to say the tournament will start at a 200-times structure for the first 4 hours and then it will change to 150 for another 4 hours and then stick at 100 for the rest of the tournament.

If you don’t like fast tournaments, my guess is that you would not enter a tournament that is a 10-times structure. You might make all your tournament “play or don’t play” decisions based upon this one “x-times” number. How simple. It would be hard for a player to complain about the tournament structure. Now, making the final table will give you a chance to maneuver if you have an average stack and will make for some great final tables.

This takes all the guesswork out of the tournament and all the complaining, except for one group of people, the companies paying for the filming. It will make more play at the final table which can increase film production costs, but oh well, the tournaments are for the players. Aren’t they?

And yes, I said get rid of the clock! (except for bathroom breaks, of course)

22 Comments to “The Blind Structure Solution”

  1. Ed

    Tom, I don’t get it. From your examples I am just not seeing how one could play the way we are use to playing now. I am short stack…do I make my move now with this “OK” hand or do I have time to wait for something better. If there is no clock to tell me I have 25 min left in this level then I have no way to judge when the blind are going up. (Unless I can judge by how many more people need to be elimated and what the stack sizes are of the short stacks.)

    Nothing wrong with a clock. Oh oh…here is a fun thing to do. The knock out button! When someone bust out they push the button and a random generator goes off and if the “BLIND CHANGE” sign lights up the blinds go up.


  2. Tom Schneider

    Thanks for the comment Ed. Your point is a good one and I have a solution. However, knowing the structure of a tounament all the way through, especially the final table, is much more important to me than knowing when the blinds are incresing when I’m short stacked.

    If you don’t like the randomness of increasing blinds then after enough people have been eliminated for the blinds to increase, the tournament director could start a clock and say that the blinds are going to increase in 10 minutes. However, one of the things I hate about tournaments is that players stall in order to make sure the blinds don’t increase on their big blind. The system that I proposed takes that kind of B.S. away from players. The increase in blinds would be totally random, as far as time is concerned.

  3. on tilt

    I like the overall concept of this for “fast” tournys, but i’m not so sure about deepstack tournys. let say you have the main event where everyone starts with 20k..blinds are 50/100. in order to move up to 100/200 and keep the same average stack means that you would have to eliminate half of the field. eliminating 4000+ people with 20k in chips at a constant blind level of 50/100 could theoretically take a week or longer if there is no catalyst to spur the action.

    i personally like deepstacks and slow blinds, but i think that would be excessive.

    just a thought

  4. Jimmy

    Great idea. Hopefully you can use your position on player’s advisory council to get some discussion of this.

  5. DanM

    I dunno, Tom, sounds nifty, kinda like a 6-max no-limit hold’em event.

    Assuming a structure doesn’t totally suck — and I define that as whenever an average stack is anywhere near a textbook short stack, i.e. 10x the big blind — then the uncertainties of a tournament’s flow is part of the fun of playing in it. If you have lots of bustouts early, for example, well that then just creates quite a few big stacks. so then the game might become slow and skillful, or things could get really wild, depending on who has aforementioned big stacks.

    you know, man, like ebbs and flows. at some times the water is rougher than at others, and the players who can figure that out have an advantage. why do you want to get rid of that?

    However, if you applied this blind “structure” to a limit event, you could let robots play and they would be near impossible to beat.

  6. Tom Schneider

    On Tilt, I agree with you, that’s why I suggested that after 4 hours of play, or so, the x-times number could change to a smaller number and then after 4 more hours to a smaller number, but at least this assures that you won’t be the chip leader with 7 times the big blind. I can’t imagine a tournament at 200 times for the whole tournament.

    Dan, I know you think uncertainties are fun, but most pros don’t like playing the hammer as hard as you. Speaking of uncertainties, what’s the over/under on number of continuous hours Pokerati will stay up and running.

  7. DanM

    Tom, you clearly have not been watching Season 4 of High Stakes Poker. But don’t worry, one of these days you will get there.

  8. The Big Randy

    Tom–shouldn’t we set the O/U in terms of minutes, not hours?

  9. DanM

    More duct-tape, please.

  10. Ed

    Loving this season of High Stakes poker and it just started.

    Dan, since TOM brought it up. What is up with the site this week? Last night it felt like you might be the victim of a DoS attack.

  11. DanM


    Death of Service … Disk Operating System … Delimiting of Selena? Sorry, Ed, I am drawing a blank on this term.

    But in a nutshell, we were attacked by evil spammers — you might have seen their 300 posts — and the process of launching our counterattack, we blew a fuse. Sorry to leave you hanging.

  12. Ed

    Just because I love you.

    I heard about the spam prob but did not see any of it.

  13. BJ Nemeth

    Tom — Sorry I’m late with the comment, but I’ve been busy covering the Borgata, and just catching up on my poker blogs/news sites today.

    I am currently working on a standardized system for judging and comparing blind structures as they currently exist (based on timed level increases), and I’m very interested in your thoughts on the subject. I’ll have more for you on that soon.

    I think your idea presented here is very intriguing, but I think it is *much* better suited for the easily automated world of online tournaments. It seems like it would be (relatively) easy to program and implement. Find a site willing to experiment, and test it out to see if the players like it.

    However, in a major live tournament, this is a drastic change in the way things are currently run. Historically, humans don’t like drastic changes unless they are accompanied by drastic benefits.

    There are also some potential downsides. Major tournaments are already 4 or 5 days long, and the poker calendar is running out of room. How long would a tournament with this structure last last? Could a tournament director guarantee that an event with this structure would end in 4 or 5 days? TV crews in particular need to know exactly when a final table will start, and most players would like to know that as well. (For scheduling flights, time off work, or whatever.)

    Realistically, this seems like a lot more work (and math) for tournament directors, so I see the chance of it being implemented in a live tournament to be near zero. If you can work to get an online site to run it as an experiment, and it works out well (attracts more players, etc.), then you’d have a shot at getting a system like this tested in a live tournament.

    But the blind structure for next year’s WSOP should definitely be a topic at the Players’ Advisory Council, and their opinions (justifiably) counts much, much more than mine. 🙂

  14. Tom Schneider

    BJ, great comments. I do think that you are right about having it work on-line first.

    I would like to address some of your concerns. First, it’s not hard to calculate when the blinds go up. Tournament Directors already adjust the number of players and the system calculates the new chip average. As soon as the average reaches a certain number, the blinds go up. No additional work. Or even more simply, you know that when you get to x number of players or tables, the avg. chip stack will be x, and therefore the blinds need to increase. Easy.

    As far as scheduling is concerned, there is already a lot of guesswork. Many tournaments at the WSOP this year were adjusted on the fly. There were a few tournaments that started as 2 day tourneys and the staff changed it to a 3-day tourney. In addition, in many of the tournaments there was discussion as to when we were going to finish and changes to the amount of time we were going to play. Often times the TD will announce that we are playing down to 27 tonight and playing down to the final table tomorrow. Nothing is changed with my recommendation.

    Thanks for you comments.

  15. DanM

    One flaw in your structure, Tom … heads-up. When down to two players, the average stack would never change.

    But maybe you like it that way? Where there’s no clock pressure when you’re down to 1-on-1?

  16. BJ Nemeth

    Good points about the WSOP prelims, and they *have* changed a few event schedules the last few years (generally moving two-day events to three days, like you said). I was thinking more about the four- or five-day events with televised final tables. (And the WSOP prelim events with scheduled ESPN final tables all started and stayed as three-day events.)

    I still think it’d be much more difficult for a TD to implement your changes than it first appears. Here are a few reasons:

    1. Tournament Clock software is pretty crappy as it is (how hard is it to show the blinds *and* the number of players remaining at the same time?), and it would require a rewrite to accommodate a blind structure based on the number of players left. I’m not saying a rewrite would be difficult, but it would require money and a competent programmer.

    2. Dealers would have to be much more aware of the changing circumstances of a tournament. Current blind levels of 60 or 90 minutes are much easier for them to track. With variable blind structures (the blinds might be constant for 3 hours, and then change 2 times in 30 minutes), I foresee a lot of mistakes regarding the size of blinds and antes.

    3. Unknown pressure on the short stacks. If you’re shortstacked, you have little knowledge of what pressure you face. Will the blinds stay constant for a while, or increase quickly? In particular, after the money bubble bursts and a flurry of players bust, it seems like the blinds would escalate quickly.

    4. It sounds like your suggestion would lead to longer tournaments. That increases the time and cost for the casino without any additional vig.

    Personally, I think “average stack size” of the entire field is a rather useless stat to an individual player. The stacks at your own table are far more relevant, as are the size of the blinds, of course. One Day 1 table might bust a lot of players early and wind up with a lot of chips in play, while another table full of tight grinders would have far fewer chips than most. Random table draws create similar discrepencies even deep in tournaments.

    There is also a big difference between the “mean average chip stack” and the “median average chip stack”. You probably already know this, but others reading this might not, so here goes a quick explanation. (Disclaimer: I am *not* a statistician. But I do know my way around chip counts.)

    When most people say “average,” they mean the “mean” average: Add up all the chips in play, and divide it by the number of players left — there’s your mean average. The median average is a different concept; it’s the point at which 50% of the field has more chips and 50% of the field has fewer chips — the player exactly in the *middle* of the chip counts.

    In a poker tournament, these numbers are usually quite different. While the mean average chip stack might be 200,000, the median average chip stack might be only 150,000 (meaning half the field has fewer than 150K).

    At most stages of the tournament, the median is well below the mean, simply on the nature of statistics. As players approach zero chips, they are very likely to bust very quickly or double up out of that range — so there are very, very few players with 10-15% of the average chip stack. However, on the high end (among the chip leaders), you are likely to survive for a long time with a chip stack at 200% of the average. That puts more chips in the hands of the leaders, and creates a field where the majority have below-average chip stacks. That creates a median average that’s well below the mean average.

    Since it would be impossible to find the median chip stack in a 100-player field (without RFID chips), you would have to rely on the theoretical mean average stack. (Once again, this could be easily computed in an online tournament.)

    In a normal tournament, a player’s relationship to the average stack (whether it’s mean average or median average) is irrelevant. The only factors at work are their stack size relative to the blinds and relative to the other stacks at his or her *table*. That’s one reason why table draw is so important, as we all know.

    In my opinion, if you don’t plan to increase the expected length of the tournament, a blind structure based on the mean average would put even more pressure on the short stacks than they currently face. And I don’t think that’s a good thing. (Everyone loves a good comeback.) If you do want to increase the expected length of the tournament, that’s great for the players, but would it be that much greater than just increasing the blinds in a “normal” time-based structure?

    Like I said, none of these issues are much of a factor in an online tournament. If a tournament runs too long, there are no additional costs. The math to compute blinds is automatic, and you could even base it on the median average chip count, if you’d like.

    If you can get an online casino to take a chance and try this, it would be an interesting experiment, and you’d probably observe things (positive and negative) that neither you nor I have predicted. I think the poker world *needs* more experimentation on things like this in order to improve the game in the long-term, and you have my full support in trying to lobby an online site to try it.

    However, if you can’t get an online casino to give it a shot, I see no chance of a live casino testing it first. It’s a pretty dramatic change, even if it’s not too difficult, and I fail to see a compelling reason from the casino’s point of view. It’ll cost the casino more time, more money, and more training for their staff. And there is little incentive to try something new like this, because it’s not like tournament attendance is disappearing.

  17. DanM

    BJ, your long-winded complex understanding of poker’s inane minutae finer intricacies makes me think we should be blogging together again … just like ol’ times. Pokerati pays about one millionth of what PokerBlog did, but hey, for you that shouldn’t matter, because it’s not like you would actually make anyone pay up.


    It’s like you do what you do for the love of the game or something.

  18. BJ Nemeth

    Funny you should mention that, Dan, because it’s still applicable. I need an accountant, or an agent, or a business manager, or something. But I don’t make enough to justify it. In my post-college life, I’ve probably earned more than $10,000 that I never invoiced.

    Perhaps if I were married, my wife could handle billing for me. But I *know* I can’t afford a wife right now. Maybe just a nosier girlfriend with too much time on her hands. Yeah, that sounds about right. I’ll get to work on that.

    My poker career reminds me of a Homer Simpson quote after his brief stint as Poochie the Dog:

    “The thing is, I lost creative control of the project. And I forgot to ask for any money. Well, live and learn.”

  19. DanM

    ***I’ve probably earned more than $10,000 that I never invoiced.***

    And the moral of that story is … never work.

  20. Tom Schneider

    BJ, I like the fact that you have responded to my post in such great detail. If I knew how to strike out words like dictator Dan, I could have written some funny things and then deleted them too.

    First, I don’t think I have ever played in a tournament that didn’t have the following info on the screen, time remaining, players remaining and average stack. Therefore, knowing when to change the blind structure would be real simple. If the tournament didn’t have that info, number of players is enough. After the buy-in window is closed, the number of players at which the blinds change, could be calculated for the whole tournament. The TD would know that when there were 79, 53, 40, 20, 15, etc. people left, the blinds are going up. Since no one would be watching the clock, they would all be focused on the number of players left. When the field got down to 80, you would know that blinds are probably going up soon.

    Second, TD’s right now, don’t know how long tournaments are going to last. They, on the fly, adjust the amount of time that will be played in order to speed up or slow down the tournament. This would be no different here.

    Third, if you want a faster tournament, you would lower the “times number”.

    Fourth, good comebacks will be possible no matter which structure is used.

    Fifth, one of the problems with poker these days, is that the customers, i.e. poker players rarely get a say in any of these types of things, because as you said, “It’s a pretty dramatic change, even if it’s not too difficult, I fail to see a compelling reason from the casino’s point of view.” There is no reason for any of the tournament providers to change much because the flocks keep coming.

    The whole idea of the suggestion is that wouldn’t it be great to know that if you have an average stack, you will never be short, unlike when I was the chip leader in one of the events at the WSOP this year I had 8 times the big blind.

    I think that implementing this structure would be simple and would at least eliminate the incessant bitching about blind structures.

    Thanks for your comments.

  21. DanM

    ***The whole idea of the suggestion is that wouldn’t it be great to know that if you have an average stack, you will never be short, unlike when I was the chip leader in one of the events at the WSOP this year I had 8 times the big blind.***

    Tom, I think that was a 7-stud event, about a third of the way through the Series? If I recall, that tournament blind structure clearly was infected with typos … or at least some flat-out miscalculations. The TDs should have assorted their dictatorial authority and announced some adjustments on as early a break as possible.

    Thanks for your comments.

  22. BJ Nemeth

    Tom, you’ve hit my points pretty well, and I hope you don’t think I’m picking on you. (I’ve commented on your most recent entry as well.) Quite the contrary, it’s nice to hear from one of the players who has put a lot of thought into these issues, rather than just serving up knee-jerk reactions.

    Like I said, I think your idea has merit, but it faces an uphill battle. Before it can be seriously considered, it needs to be tested in a real-world scenario, and I think we both agree that is most likely to happen online. So I’m going to stop arguing with you on this and offer my help (however you may need it) to get an online poker site to give it a shot. I think an online casino *would* have something to gain (publicity, curious players) from an experiment like this. And if it goes over well and the players like it, that same casino could attract more online players with tournaments in the new format.

    There isn’t a lot of competition in high-stakes live poker tournaments, because the calendar hasn’t been completely filled yet. (But it’s getting close.) Tournaments with crappy blind structures can still attract hundreds of players. Tournaments that charge a lot of vig can still attract hundreds of players. Live tournaments have little reason to innovate in that environment. (Innovation = Risk)

    Online casinos face much, MUCH more competition, so innovations like this are more likely to be rewarded.