Tourney Increasing "play" in the endgame (1 Viewer)

Mr Winberg

Full House
Joined
Jan 10, 2019
Messages
3,070
Reaction score
7,687
Location
Sweden
My first real post! Hoping to get some thoughts from all the experts here. :)

Most tournaments turn into shove fests during the endgame. The knockouts simply don't keep up with the increasing blinds, and eventually the average stack becomes too small for any real play. A way to solve this is to have longer tournaments, but what if you're maxed out in that departement?

What tips do you guys have for increasing the "play" (for lack of a better word) in the mid and endgame, i.e., increasing the nr of big blinds in the average stack? I've tried a few things (not sure if all of them work, though, but I've seen some improvement overall), but would like to hear from you before I share. :)

Also, which parameters would you say affect this (apart from duration)? My guess would be the nr of hands per hour, the players' incentive to play hands, and the tournament structure.

Thanks!
 
Well well well... seems like my debut topic on PCF wasn't as spectacularly interesting as I had hoped! :ROFL: :ROFLMAO:

Well, for me it's interesting, since I feel like I've failed as a host if everyone at the final table is short stacked.

I'll give an example of one of the things I've tried which I feel has worked:

One thing I started with some time ago, was to only have 8 per table. I actually prefer 10 handed, but reducing to eight allows for more hands per hour and more "hand involvement", both of which slightly increase the knockout rate (and therefore the average stacks).

For me, eight is a sweet spot in that the speed of play increases but it's still not a true shorthanded tournament.

As always, I'm all ears and looking for tips to improve :)
 
I think your move to 8 is a great choice. Have you posted your blind structure at all? That can have a big impact if you’re jumping too fast.

The one thing you can’t control is how people play short stacks. Many of the better tournament players know they need to make moves before they get under 10BBs or so, but that’s not often the case with rec players. If you get a handful of players that basically let themselves get completely blinded out, it effects the play of the rest of the table as middling stacks don’t want to get knocked out before one of the micro-stacks. So everyone tightens up until the shorties are gone, but now you’re a couple blind levels deeper and the play can’t open up as much.

What other adjustments have you tried?
 
Tournaments, by design, come to a conclusion at a certain level. That level is dependent upon the number of chips in play. It simply happens, you cannot avoid it, and you dont want to. It's how you can assure a certain amount of play time, and assures the tournament won't last all night.

That level, for what it's worth is C/20=B. C= chips in play, B = the Big blind at the tournament's end.
 
A few couple things come to mind that might be places to look.
  • Blind Structure - like others mentioned, time and blind jumps will make a difference.
  • Second deck - always having a deck ready to go after a hand really speeds things up.
    • We have the dealer shuffle and pass to the SB when the hand is over.
  • Chip breakdown - adequate number of chips? Making change wastes time.
  • It's unconventional and a bit of a pain, but you could always add time to the levels at a predetermined number of players, or at a certain level.
That level, for what it's worth is C/20=B. C= chips in play, B = the Big blind at the tournament's end.
In other words, when there are 20 big blinds in play, the tournament should be ending or over. Another shorthand would be when the big blind is equal to 5% of the chips in play.

Here is a sample structure for you to compare against. This would be a 10 players, 20 Minute levels, T2500 structure.

LevelSBBB7:00 PM StartBB Incr %BB's in PlayBB % of Chips
110207:20:00 PM-12500.1%
215307:40:00 PM50.00%8330.1%
320408:00:00 PM33.33%6250.2%
430608:20:00 PM50.00%4170.2%
540809:00:00 PM33.33%3130.3%
6601209:20:00 PM50.00%2080.5%
7801609:40:00 PM33.33%1560.6%
810020010:10:00 PM25.00%1250.8%
915030010:30:00 PM50.00%831.2%
1020040010:50:00 PM33.33%631.6%
1130060011:20:00 PM50.00%422.4% - USUALLY ENDS HERE
1240080011:40:00 PM33.33%313.2% - OR HERE
13600120012:00:00 AM50.00%214.8% - ENDS HERE FOR SURE
14800160012:30:00 AM33.33%166.4%
151200240012:50:00 AM50.00%109.6%
16160032001:10:00 AM33.33%812.8%
 
Illustrates the point well, but ... starting at 10/20 with 125 BB stacks? Strange on both counts... :confused

Oh, yes forgot to mention 500 of the 2500 is the on time bonus.

We usually play T1200 + 300 and start at 5/10 but I didn't want to post that structure which has a 100% initial increase that is upsetting to some around here. :)
 
Sorry for wall of text, but there were so many great responses!

I think your move to 8 is a great choice.
Thanks! Yes, I find the tournaments more enjoyable because of the reasons I mentioned in #2

Have you posted your blind structure at all?
No, not yet. The reason is that I've tried a few different, quite standard, none are really controversial (except one, but I'll get back to that...), some even follow all of BGinGA's rules. :)

The one thing you can’t control is how people play short stacks. Many of the better tournament players know they need to make moves before they get under 10BBs or so, but that’s not often the case with rec players. If you get a handful of players that basically let themselves get completely blinded out, it effects the play of the rest of the table as middling stacks don’t want to get knocked out before one of the micro-stacks. So everyone tightens up until the shorties are gone, but now you’re a couple blind levels deeper and the play can’t open up as much.
I totally agree. Nits nurturing short stacks can really destroy the endgame. Beyond the reasons you mentioned, having too many short stacks at a table affects the few who have decent stacks too, because unless they get in a hand with each other, it's still all-in-or-fold poker.

Second deck - always having a deck ready to go after a hand really speeds things up.
Yes, yes, yes! (@WedgeRock , do your job!) I started with this many years ago, and will NEVER go back to one deck!
Chip breakdown - adequate number of chips?
ehhh...probably the opposite...I could increase speed with fewer chips, but...what's the saying? MOAR chips!

Tournaments, by design, come to a conclusion at a certain level. That level is dependent upon the number of chips in play. It simply happens, you cannot avoid it, and you dont want to.
That level, for what it's worth is C/20=B. C= chips in play, B = the Big blind at the tournament's end.
I've seen this rule in many forms, all having the same math behind it. I view it as "when there's 20 big blinds left in play...", which is equivalent with what you wrote. But, my view of the logic behind it (and also my experience with it) is that if the tournament will last no later than that level (because at most the players have 10 BBs each). But, if you and I were the final two players with, say, 50 BBs each, I'm pretty sure the tournament would end earlier. You'd probably beat me within a level or two! ;-)

Before I started trying to improve this stuff, my tournaments usually ended with 20-30 BBs left. Out of the last 5 tournaments I've held, four ended with 30-50 BBs left. The fifth was my fault, I was heads up against a player who had no experience playing heads up and was super tight, so I played small ball for 3 levels, slowly grinding the player down until there was nothing left. Tourney ended with ~25 BBs left in play.

What other adjustments have you tried?

This post is long as it is ;-) but ok, one more. This one is perhaps marginal, but I think it gets us down to 7 players a level earlier than we otherwise would, meaning more "play" for them:

It's the number of players at the final table. In my experience, the slowest play of the tournament (apart from maybe the bubble play) is when the final table has formed. The blinds are already high and the decisions tough. Having a full table make the play really slow. Therefore, I don't have 8 seats at the final table, I have 7. So, when 8 are left, instead of having 8 players nit it out on the same table, there are two action packed shorthanded tables of 4 each (with more hands per hour, and more incentive to get involved). We will sooner get down to seven than we would do if we have 8 at the FT. I got the idea at a friends tourney with 10 at the final table. We didn't make an orbit for two levels with all the tanking that was taking place. People who don't like short handed play will of course hate this idea...
 
My first real post! Hoping to get some thoughts from all the experts here. :)

Most tournaments turn into shove fests during the endgame. The knockouts simply don't keep up with the increasing blinds, and eventually the average stack becomes too small for any real play. A way to solve this is to have longer tournaments, but what if you're maxed out in that departement?

What tips do you guys have for increasing the "play" (for lack of a better word) in the mid and endgame, i.e., increasing the nr of big blinds in the average stack? I've tried a few things (not sure if all of them work, though, but I've seen some improvement overall), but would like to hear from you before I share. :)

Also, which parameters would you say affect this (apart from duration)? My guess would be the nr of hands per hour, the players' incentive to play hands, and the tournament structure.

Thanks!

When you say "most of", what number are you talking about? Small stacks are sticking around won't happen every time. Your structure may be just fine. I'd look at at least 8-10 tourneys before you mess with stuff.
 
Having a table with fewer players that your normal play goes 100% against TDA rules, which suggests the same number of players, +1 (but no more than 10). So if you have 2 tables of 8, your final table should have 9 players, if seating is possible.

From a recreational point of view, getting all the players at a single table is far more preferential. It's a social game. Keeping everyone split up is...anti-social.

But, if you and I were the final two players with, say, 50 BBs each, I'm pretty sure the tournament would end earlier. You'd probably beat me within a level or two! ;-)
If you and I have 50 BB each, we have a ton of time (depending on the levels). When the average stack dwindles to 10BB, by all reasonable strategies it will be a shove-fest. However, nits can continue to nit it up. It's called Playing the game. Not every player plays "Win or go home" strategy. Some are happy to limp into the money and double up. That's how they play, and it's a valid (albeit -EV) strategy.

You blame the nits for protecting their stack, wanting short-stacks to shove. Then you say you're opposed to shove-fests. You can't have it both ways.

In the end, nearly everyone will have to shove at some point. You set the tournament with the blinds, to give X number of hours to play. They can get shovy with 20 BB, 12 BB, or 3 BB. It's all up to them. Your blinds won't change that.
 
Most tournaments turn into shove fests during the endgame.
Does this happen before the money bubble or after? Once the last X (3, 4, 5,?) players are in the money, you could write the rules so that the duration of the level at which the money bubble bursts gets extended, for the remaining players in the money to have more hands at smaller blinds for an extended period of time.

I've played in some home game tournaments where once you made the money, the blinds went up so quickly after that, that it became a shovefest. (Some were designed like that because the host needed to end the tournament at a certain time). But if time wasn't a constraint, I always thought that extending the level after the money bubble burst would allow for some more poker strategic poker play for the people making the money before it turned into a shove-fest.
 
When you say "most of", what number are you talking about? Small stacks are sticking around won't happen every time. Your structure may be just fine. I'd look at at least 8-10 tourneys before you mess with stuff.
Yeah, it took about 8-10 tournaments before I started messing about, actually :)

Having a table with fewer players that your normal play goes 100% against TDA rules
I know. Guilty. As I said, this has marginal effect, probably gets us down to seven a level earlier. I'm not totally married to it. It hasn't had as big effect as going from 10 to 8 players.

From a recreational point of view, getting all the players at a single table is far more preferential. It's a social game. Keeping everyone split up is...anti-social.
You're right. The problem is, the more social we get, the slower the less experienced players get. But it is a trade-off, for sure. I wanna have fun, both socializing AND check-raising the flop without being pot-committed, but I can't have both.

You blame the nits for protecting their stack, wanting short-stacks to shove. Then you say you're opposed to shove-fests. You can't have it both ways.
I might have used the word shove fest wrong. What I meant was having the average stack so low that people may as well shove when re-raising preflop, or even on the first raise.

I have had (probably through luck and coincidences) tournaments where there actually was some play all the way to the end. Other times, the average stack gets too low for allowing play (what I was trying to refer to as "shove fests"). What I'm trying to achieve is to have more of the former, and less of the latter, if that is at all possible without extending the tournament.

Does this happen before the money bubble or after?
It's probably worse before the bubble, but it's more in general when the blinds get higher. It's not that I think my tournaments are worse than other's, it's just that I want to improve them in this aspect.

But if time wasn't a constraint, I always thought that extending the level after the money bubble burst would allow for some more poker strategic poker play
It's a cool idea, but time is the real issue, isn't it :)
I would very likely have a lot of strategic play throughout the tournament if I could use the main event's structure and blind levels (and dealers). The only way I can think of making the blind levels longer without making them longer is increasing the nr of hands per hour, which is one of the areas I'm looking for advice on.

One thing I've tried which I think works relatively well is placing flyers on each table with the structure for the tournament, and any other information I would otherwise have to say again and again... This allows for a slightly longer tournament. On the flyers I've written tips for speeding the game up. Those who read it learn something and speed up the game. Those who don't at least know that the host cares about speed. Since I started this, I've found that people pay a little bit more attention. (it might be that they've become more experienced, though)
 
A fellow Scandinavian. :) Yes, seems like similar topics.

I see you mentioned in your thread having only a few people dealing. A helpful member actually gave me the same advice via PM, so that is something I might try. I don't know how to solve the logistics, though. We're usually 3 tables, that means either 3 or 6 dealers (depending if they alternate). It wouldn't be fair to allocate the best dealers to different tables before the table draw, as these are also the best players. Unless I call it "seeding" ;-)
 
A fellow Scandinavian. :) Yes, seems like similar topics.

I see you mentioned in your thread having only a few people dealing. A helpful member actually gave me the same advice via PM, so that is something I might try. I don't know how to solve the logistics, though. We're usually 3 tables, that means either 3 or 6 dealers (depending if they alternate). It wouldn't be fair to allocate the best dealers to different tables before the table draw, as these are also the best players. Unless I call it "seeding" ;-)

At my events I will have the players draw from one set of seating chips and the dealers pull from another. So yes the dealers are still all split up but it’s at least somewhat random what table they deal.

Sure it’s not 100% random but it’s the only alternative I can think of. Outside of hiring non-playing dealers which would cut into the prize pool and starts to border on illegal depending on local gaming laws.
 
Hey, got your message in the other thread. Yeah, there's definitely a give and take to it all. Here's a couple things I learned:

Check out the blind schedules Matt Savage made for the WPT. A quick google and you'll find one he made for the LA Poker Classic. I'm a big fan of that structure. It uses a T25 base. The ante forces the shorties into action, so you have fewer people hanging around, diluting the average stack size. I think that's the single most important thing you can do. The beauty of it is that the blinds double every 3 levels. Its a nice even pace. Any faster than that, especially with short levels, and the blinds race up on you pretty quick.

Imagine this. Your home game has 20 minute levels, self-deal. At your table is GUY WHO HAS TO BE REMINDED TO ANTE EVERY HAND, next to GUY WHO DRANK TOO MUCH, across from GUY WHO HAS TO STAND UP TO DEAL. Pace of play is pretty slow, so if you're lucky you're getting in one orbit a round. Now, its crunch time, 2000/4000 and the bubble is coming up. Average stack is around 30BB. If the blinds go from 2K/4K to 3K/6K to 4K/8K, they double in the course of two rounds. Theoretically, you can go from the end of 2/4 to the start of 4/8 in the course of 7 or 8 hands tops. That's not enough time to lose players and now the avg stack is 15BB. Shovefest time.

If you know how many players you have, you know how many BB are in play at any given time. Rule of thumb; no matter how deep you start, the blinds won't start forcing people out until the avg starting stack is about 50BB. Sure you may lose a couple before that, but it won't be because of the blinds, and if you allow reentry they are likely to come back in, so your field doesnt really shrink much until then. From there, you're going to hit a point where the average stack stays fairly consistent -- and instead of the avg stack shrinking, the field shrinks. For my tourney (with an ante and the Savage structure) that number was about 30BB. And thats where the avg stack stayed until we were short handed. If you don't use antes or start skipping levels, that avg stack constant will be smaller (maybe 15-25 ish).

Adding chips then doesn't change the pace at which you lose players, only the point in time at which the pressure kicks in. So, if the blinds double every 3 levels, an you double the starting chips, you've just added 3 levels to the beginning, before people will feel the heat. Good way to make sure everyone has at least a little time at the table, and "gets their money's worth." Obviously, there's a balancing act between losing people gradually and making sure everyone has enough of a good time. I hate a shovefest, but I also hate being blinded out before the first break.

Believe it or not, I'm a bit of an analyst by trade, and I was able to build a model in Excel that took the blind structure and inputs for players, starting stacks, and a prediction on average stack size and was able to model how many people I'd have left during any given round. It wasn't far off. It made it easy to play around with What If scenarios, like what happens if I add or subtract players or starting stacks. If you think something like that would help, DM me your email and I'll send you a copy.
 
Hey @Prenders17, thanks for a very detailed post! I really appreciate it.

First, regarding antes, short story there is that I agree with you, and I started using antes several years ago for the reasons you bring up, and I experienced a positive effect. I've switched over to using BB antes now for convenience.

However, I hope this thread doesn't turn into a war between antes vs no antes, or worse, BB ante vs standard ante. Therefore I want to be very clear that I totally respect that antes are not for everyone, and that it's perfectly fine to run home games without antes. :tup:

Using a structure that is that smooth (i.e., doubles every third level) kinda necessitates that you don't start too deep. After a quick google search for WPT's structure (this is what I found: https://wptmedia.wpt.com/wp-content/uploads/2014/12/WPT_SpringChampionship_STRUCTURE.pdf, which looks very similar to WSOP's), I find that if I want 20 minute levels, I would have to start my T20000, 5 hour playtime (excluding breaks), ~20 player tournaments at the level where the BB is 500! (My tourneys usually end when there's 50-30 BBs left, usually closer to 30, so that means a BB of ~12k)
To start at a more reasonable 150 means I need to drop down the level time to 15 minutes.

Would you recommend skipping a few levels and have a reasonable level length, or have short levels?

I'll totally take you up on your offer regarding the spreadsheet! Thanks!
 
One idea I always wanted to try is starting at level 3, and with the freed up space lengthening the rest of the levels or adding two levels to slow the pace of the blinds. The first couple levels are fun, sure, but tourney time is precious and do I want to burn 30-40 minutes there to have the real action rushed later on?

Just a thought. Haven’t tried it in practice but if you give it a go let me know how it turns out.
 
You will always have shovefests at the end unless your tournament has very long levels & overall run time. Even then, you could end up with a few tight players who don’t have any major coolers that play until their stacks are average 10bbs. The issue is with shorter tournaments there just aren’t enough hands to have opportunities for big coolers/hands to knock out enough folks. It’s up to the blind increases to force folks to start shoving and going all in. If you want more play in your endgame, extend your level time or add additional levels, but even that won’t work 100% of the time.
 
One idea I always wanted to try is starting at level 3, and with the freed up space lengthening the rest of the levels or adding two levels to slow the pace of the blinds.
Yes, I think this is in line with what @Prenders17 is saying. The question is how far you want to push the idea? Perhaps a good compromise is to keep the early levels, but shorten them? That way everyone gets their warm-up levels and can enjoy some deep stack play, but more time is allocated to the mid and end game.

if you give it a go let me know how it turns out.
Dito! :tup:
 
You will always have shovefests at the end unless your tournament has very long levels & overall run time.
I agree, except that I think (hope) that it's possible to influence this, to make it less shovey and more playey.

If you want more play in your endgame, extend your level time or add additional levels, but even that won’t work 100% of the time.
Yes, that is a good solution, if you can. Most of the time you have a certain time frame to work with, and for me (currently) that is about 5 1/2 hours from the doors open until the final hand, of which 30 minutes are needed for breaks and getting everyone seated. So the question in the OP is basically how to make the best tournament possible given a certain time frame.
 
Around 20 entrants +/- 6, usually 5 hours effective time excluding breaks. I start with 200 BBs, sometimes less. I usually run T10K or T20K, but have thrown in T15K and T25K as well. The foundation of my blinds is this
50 (10K starts here)
100 (20K starts here...)
150 (...or here)
200
300
400
600
800
1200
1600
2000
3000
4000
6000
8000
12K
16K
20K
but I play around with different tweaks, so I don't follow it slavishly. They usually end when or before there are ~30 BBs left. The "record" is ending with 60BBs left (and it was the first hand on that level!) but that was largely due to a massive chip leader keeping all other stacks low.

If following your advice, @Prenders17, I should start a bit later (and/or shorten the early levels) and add more levels later on, like all or some of 500, 1000, 2400, and 5000, do I understand correctly?
 
Adding those 250/500, 500/1000, 1200/2400 and 5K/10K multiples will increase play in the mid and late stages. Obviously, that adds duration, and if you have a 5 hour hard stop, trade offs have to be made. I definitely think 10K is the right starting stack. Deeper is nice if you have the time, but if you don’t then you’re just compressing the mid-stage. Let me ask, when you do 10K chips, do you ever go long/finish early with 20 guys? What about at 26 guys?
 
I definitely think 10K is the right starting stack. Deeper is nice if you have the time, but if you don’t then you’re just compressing the mid-stage. Let me ask, when you do 10K chips, do you ever
T10K vs T20K is more about variation than about depth since I adapt the starting blinds accordingly. For example, T20K is actually less deep when I start at 150.

Let me ask, when you do 10K chips, do you ever go long/finish early with 20 guys? What about at 26 guys?
We usually don't go long, since I calculate the level lengths based on total chips in play and use "30 BBs in play" to see where it will end. After this level, I increase the jumps and shorten the levels to make sure that it doesn't end too late, but we very seldom reach past 30 BBs in play. To answer your question, I can't say for sure, but it feels like we more often finish early with fewer players than with many players, even though the structure is adapted accordingly.
 
Just sent you that file, with the modifications I might make. But of course, its your show, so play around with it at your leisure. Good luck.
 

Create an account or login to comment

You must be a member in order to leave a comment

Create account

Create an account and join our community. It's easy!

Log in

Already have an account? Log in here.

Back
Top Bottom