Tourney new idea non standard for blind increases (1 Viewer)

Joined
Nov 17, 2014
Messages
160
Reaction score
157
Location
USA
I was thinking about the first 2 blind levels in conventional T10k structure and how many start with 25/50 and then double to 50/100. Some, like BGinga, suggest starting at 50/100 and then go to 75/150 to lower the delta. This makes sense but if you want 200BB stacks then you need to increase chips to T20k. This means I would need to have more chips and also starting BB is 2 chips. Not a big deal really, but then I started to think about alternative to buying more chips.

What are your thoughts on starting at 25/50 and then next level is something non-standard like 25/75? Practically, I don't think the SB amount matters at all in betting. In fact, you see this a lot in cash: $0.25/$0.25, $1/$1, $1/$3, $2/$5. This would allow you to have T10k tourney, start with 25/50 for 200BB stacks, and have smoother transition to next level.

Thoughts?
 
Last edited:
This is better than having the blinds double, but I just don't see what's wrong with 20K starting stacks. Just add two 5K chips to your 10K standard starting stack.
 
I really don't get all the fuss about blinds doubling from 25/50 to 50/100 if they are the first 2 levels. *shoulder shrug*
 
That's doubling the BB, so same issue.

Plus, you'd still have the same problem after 25/50, just at the third level instead of the 2nd. Like links, I just don't see this is a big deal at this level.
 
That's doubling the BB, so same issue.

Your making an issue out of a non-issue. If we were talking about blinds doubling later in the tournament - like 500/1000 -> 1000/2000 -> 2000/4000 - I would be singing a different tune. If the first four levels are 25/25, 25/50, 50/100, 75/150 you'll be fine.
 
Your making an issue out of a non-issue. If we were talking about blinds doubling later in the tournament - like 500/1000 -> 1000/2000 -> 2000/4000 - I would be singing a different tune. If the first four levels are 25/25, 25/50, 50/100, 75/150 you'll be fine.

I concur.

Sal,

How long are your blind levels? I know a lot of home structures start out at say 30 minutes, then drop to 20 after 3-5 levels. If you want more poker playing later in the game try starting the first three or four levels at 20 minutes, then move to 30 minute rounds. We made this change a couple of years ago and its a huge improvement imo.
 
I'm not trying to make an issue out of it, I just had an out-of-the-box moment and wanted to hear thoughts on it.

What inherently makes doubling bad later in tourney rather than earlier? Most discussions talk about smooth transitions between blind levels so I am just throwing an idea out there on how to avoid it for any starting size.
 
WSOP uses a 25/25 then progresses to 25/50. Technically the bb doubles, but the SB remains the same, so the cost per orbit does not double in the first 2 blinds. It does double from lvl 2 to 3, but again, at such a low level, it's a non issue.

That said, for years I've used a 700/1500 blind level. To me, it was better than working all the hundos of a 700/1400 (min raise to 2800). I've only heard one player comment on in, and it was not a negative comment, just a "hey, that's not half". If you wanted a 25/75, it wouldn't shake my tree.
 
It may seem insignificant, but doubling the blinds at any time during a tournament has a huge effect. Please note that I said doubling the blinds, not doubling the big blind. There is a huge difference.


Let's look at the beginning levels of several different tournament structures, each of which features ~200BB starting stacks. For ease of comparison, all will assume 10 players, 15-minute blind levels, and no antes, but the conclusions will hold true for any number of players or blind level lengths. Adding antes (pro/con/size/when) is another topic outside the scope of this post, and will not be addressed here.


Before we get started, let's also look at how the number of total blinds (big blind plus small blind) relates to the starting stack size (important because total cost of the blinds is more relevant to stack size than just the cost of the big blind):

- For starting blinds of 25/50, a 10K starting stack contains 200 big blinds or 133 total blinds (BB+SB).
- For starting blinds of 50/100, a 20K starting stack contains 200 big blinds or 133 total blinds (BB+SB) -- the exact same relationship as 25/50 blinds with a 10K starting stack.
- For starting blinds of 25/25, a 5K starting stack contains 200 big blinds but only 100 total blinds (BB+SB). Even though the number of big blinds is the same as the two previous structures, this one will not play nearly as 'deep'.
- For starting blinds of 25/75, a 15K starting stack contains 200 big blinds and 150 total blinds (BB+SB). In essence, the 15K stack at 25/75 is 'deeper' than any of the other structures, when taking into account the actual total blinds represented by each.

Of the four choices above, 15K stacks with 25/75 starting blinds allow players more overall maneuverability and deeper play, even though the starting stacks in each example all contain 200BB.


When comparing the different structures below, we will be tracking the following:
- bbp: the number of big blinds per player at any given point in the tournament
- abp: the number of all blinds (BB+SB) per player at any given point in the tournament
- bbt: the total number of big blinds in play at any given point in the tournament
- abt: the total number of all blinds (BB+SB) in play at any given point in the tournament

In addition, we will also track:
- %bb: the percentage increase in the big blind from one level to the next
- %ab: the percentage increase in all blinds (BB+SB) from one level to the nextal
- %avg: the average increase in blinds at any given point in the tournament


Lastly, before we begin, let's look at how asymmetrical small blind amounts (those that are not 50% of the big blind size) interact and can affect things like the average blind increase and total number of blinds per player. A typical blind structure might look like this:
sb bb %bb %ab
L1 50 100 0 0
L2 75 150 50% 50%
L3 100 200 33% 33%
L4 150 300 50% 50%


but by changing the relational values of the small blinds, we can still mimic the 50-33-50 blind progression using smaller starting blind amounts:
sb bb %bb %ab
L1 25 25 0 0
L2 25 50 100% 50%
L3 25 75 50% 33%
L4 50 100 33% 50%


Note that even though the big blind percentage increases differ, the percentages of total blind increases (%ab) are identical in both structures. In addition, note how it can also be acheived using small blind amounts that are equal to the big blind:
sb bb %bb %ab
L1 25 25 0 0
L2 25 50 100% 50%
L3 50 50 0% 33%
L4 50 100 100% 50%


Again, even though the big blind increases vary widely, the percentages of total blind increases (%ab) are identical.


So what's the best structure to use? Well, that depends a lot on the availability of chips, time, skill, and patience.


Each structure below lists the first six blind levels (totaling 90 minutes of play), displaying the small blind, big blind, number of big blinds per player (bbp), number of all blinds per player (abp), total number of big blinds in play (bbt), total number of all blinds in play (abt), percentage increase of the blinds - big blind (%bb) and total blinds (%ab), plus the average increase in blinds for the tournament at each of the given blind levels.

Structure #1 (s1) - 10K stacks and 25/50 opening blinds:
sb bb %bb %ab bbp abp bbt abt %avg
L1 25 50 0 0 200 133 2000 1333 0.0%
L2 50 100 100% 100% 100 67 1000 667 100%
L3 75 150 50% 50% 67 44 667 444 75%
L4 100 200 33% 33% 50 33 500 333 61%
L5 150 300 50% 50% 33 22 333 222 58%
L6 200 400 33% 33% 25 17 250 167 53%

Players drop from 200BB to 100BB immediately (or 133ab to only 67ab), and are shortstacked by the end of L6.

Structure #2 (s2) - 20K stacks and 50/100 opening blinds:
sb bb %bb %ab bbp abp bbt abt %avg
L1 50 100 0 0 200 133 2000 1333 0.0%
L2 75 150 50% 50% 133 89 1333 889 50%
L3 100 200 33% 33% 100 67 1000 667 42%
L4 150 300 50% 50% 67 44 667 444 44%
L5 200 400 33% 33% 50 33 500 333 42%
L6 300 600 50% 50% 33 22 333 222 43%

Superior to s1, this structure allows players more depth at all levels and has both a more uniform and lower average blind increase.

Structure #3 (s3) - 10K stacks and 25/50 opening blinds with one asymmetrical SB amount (25/75):
sb bb %bb %ab bbp abp bbt abt %avg
L1 25 50 0 0 200 133 2000 1333 0.0%
L2 25 75 50% 33% 133 100 1333 1000 33%
L3 50 100 33% 50% 100 67 1000 667 42%
L4 75 150 50% 50% 67 44 667 444 44%
L5 100 200 33% 33% 50 33 500 333 42%
L6 150 300 50% 50% 33 22 333 222 43%

Also superior to s1, this structure mimics the progression of s2, but actually improves playability during L2 (100abp vs 89abp with s2). Note that this can also be accomplished using 50/50 blinds during L2 in lieu of 25/75:
sb bb %bb %ab bbp abp bbt abt %avg
L1 25 50 0 0 200 133 2000 1333 0.0%
L2 50 50 0% 33% 200 100 2000 1000 33%
L3 50 100 100% 50% 100 67 1000 667 42%


Structure #4 (s4) - 15K stacks and 25/75 opening blinds with one asymmetrical SB amount (25/75):
sb bb %bb %ab bbp abp bbt abt %avg
L1 25 75 0 0 200 150 2000 1500 0.0%
L2 50 100 33% 50% 150 100 1500 1000 50%
L3 75 150 50% 50% 100 67 1000 667 50%
L4 100 200 33% 33% 75 50 750 500 44%
L5 150 300 50% 50% 50 33 500 333 46%
L6 200 400 33% 33% 38 25 375 250 43%

Superior to all three previous structures -- it plays deeper initially due to a higher total starting blinds per player, but also plays deeper at L4 (50abp vs 44 with s2/s3, or just 33 with s1) and L6 - plus this expands into the later levels (L7+) as well.


Overall, the 20K structure (s2) has the lowest average blind increase over the life of the event (39%), vs 41% with S3/s4 vs 43% with s1. The s1 structure will finish sooner, but if a shorter duration tournament is desired, it's better to shorten the blind level times of a better schedule than to compromise the integrity of the event by doubling the blinds.

And there's certainly nothing wrong with using a 25/75 blind level when used properly, imo.




- - - - - - - - - Updated - - - - - - - - -


Some may have taken a glance at my post above and thought, "Holy crap -- why is ANY of that stuff important?"


Well, here's why. The advantage of deep-stack tournaments is not that players start with huge amounts of chips. The real advantage comes from still having large amounts of chips (relative to the blinds) later in the tournament -- and not just for a couple of levels, either.


For example, it's pretty senseless to start with 1000BB when the blinds double every round. Players end up at the same place after six levels as if they had started with 200BB using a structure with 40% increases, but most of those early levels were at stakes so low that very little chip movement occurs between players -- making them essentially a waste of time. Worse, as blinds continue to double, the event quickly escalates into a shove-fest as nearly everybody becomes simultaneously short-stacked:


L1 - 1000BB
L2 - 500BB
L3 - 250BB
L4 - 125BB
L5 - 62BB
L6 - 31BB
L7 - 15BB
L8 - 7BB

That's the extreme end of the scale, but lessons are there that apply to more reasonably-fashioned blind structures as well. As stated earlier, it may seem insignificant, but doubling the blinds at any time during a tournament has a huge effect.

The advantage of 100BB to 200BB stack tournaments where blinds do NOT double is that the early rounds are all played with stacks that contain a meaningful number of big blinds, and those stacks continue to be meaningful for a reasonable amount of time before becoming short-stacked. Where exactly is that sweet spot? With NLHE, it is typically stacks that range between 40BB and 150BB where the most skillful play is both allowed and required.
 
Last edited:
Thanks for the post BG. ABP as the key metric for how short stacked (and therefore less skillful) the game gets makes sense to me. The thing I'm not sold on is that awkward (in terms of chip requirements) blind amounts are the only or best way to slow and smooth the ABP decline. Obviously for fixed length levels, larger blind jumps necessarily shift the ABP/time curve down vs. smaller jumps, but by varying the minutes per level, the tournament can end up at the same average ABP/minute (with fewer levels) over the first two hours, for example. Is there an argument I'm missing against that train of thought?

Also, how would a structure #5 that added an initial 25/25 level to structure #1 (or a better structure) stack up in your comparison? (Reading more carefully, I'm guessing it's a bit of a time waste that doesn't promote stack size separation. For a friendly game though, I can see the benefits of having some cheap social hands for less serious players.)
 
Last edited:
Thanks for the post BG. ABP as the key metric for how short stacked (and therefore less skillful) the game gets makes sense to me. The thing I'm not sold on is that awkward (in terms of chip requirements) blind amounts are the only or best way to slow and smooth the ABP decline. Obviously for fixed length levels, larger blind jumps necessarily shift the ABP/time curve down vs. smaller jumps, but by varying the minutes per level, the tournament can end up at the same average ABP/minute (with fewer levels) over the first two hours, for example. Is there an argument I'm missing against that train of thought?

I have long been a proponent of altering tournament blind level lengths so that the same (or similar) number of hands are played at each blind level. With fixed blind times, fewer hands per hour are played at a full table vs a 5- or 6-handed table, for example. Balancing the hands/hour throughout the tournament is relatively simple for a single table (blind levels get shorter as the table gets smaller), but it is more difficult to implement with bigger events where players get moved to balance tables and when tables get collapsed and combined (going from fewer player back to full capacity).

Another important thing to note: any tournament structure (no matter how well-designed or ill-advised) has an optimum play strategy. Players who can identify and utilize those strategies based on current conditions are way ahead of the pack, vs those players who sit down to play with no knowledge of the blind structure or a plan based on that knowledge.

Also, how would a structure #5 that added an initial 25/25 level to structure #1 (or a better structure) stack up in your comparison? (Reading more carefully, I'm guessing it's a bit of a time waste that doesn't promote stack size separation. For a friendly game though, I can see the benefits of having some cheap social hands for less serious players.)

Here is such a structure (T5000 starting stacks = 200BB or 100ABP), which can be compared to those in the previous post:

Structure #5 (s5) - 5K stacks and 25/25 opening blinds with two asymmetrical SB amount (5/25, 25/75):
sb bb %bb %ab bbp abp bbt abt %avg
L1 25 25 0 0 200 100 2000 1000 0.0%
L2 25 50 100% 50% 100 67 1000 667 50%
L3 25 75 50% 33% 67 50 667 500 42%
L4 50 100 33% 50% 50 33 500 333 44%
L5 75 150 50% 50% 33 22 333 222 46%
L6 100 200 33% 33% 25 17 250 167 43%

Note that to properly compare all of the example structures, the starting stacks should be altered so that they all contain the same number of total blinds, not big blinds.
 
Last edited:
Another important thing to note: any tournament structure (no matter how well-designed or ill-advised) has an optimum play strategy. Players who can identify and utilize those strategies based on current conditions are way ahead of the pack, vs those players who sit down to play with no knowledge of the blind structure or a plan based on that knowledge.

^^^ This.

Before I play a tournament, I use a spreadsheet to input blind structures and determine ahead of time how much patience I'm allowed, thus determining my hand range. It frustrates me to no end when casino's poker rooms have websites speaking of great tournaments but not posting blind structures, or equally useless, just the first blind level. I need to know length of time for each level and the SB/BB/ante. Approx number of players is also helpful, but sometimes getting this info is like pulling teeth from an angry bear.

If there was only some way casinos knew how often I skip playing in their tournaments just because this info was not available online.
 
BG, your explanations above are excellent. Like Zombie, I input info into a spreadsheet to decide if I even want to play in an event. Based on your info above, I just added 3 columns to the spreadsheet I use.

I host a monthly game. Our group likes deep stacks, and wants to be finished in 4 hours. The basic formulas of a tournament will end when the BB = 5% of the total chips in play or when the SB + BB + Antes = 5% of the total chips in play help determine when a tournament will end. My goal is to keep 90% of the players in for the first 2 hours (through 6 blind levels for us using 20 min blinds), but end between 3:40 and 4:20. I've not asked the group, but for me, I want a tournament where the better players have a better chance to win. The info you provided can help someone make their tournament better.

I do agree that doubling blinds is not ideal, but between the first and second round, it has an advantage. It rewards the players who show up on time. We give players a full stack when they come, but it's really only worth half as much if they miss the first blind level. For everyone who is there on time, that first round usually is not real significant unless two strong hands get into it, which could happen any time with any structure.

Here's another thought to stair step your way up starting at 25/50:
Round SB BB
R1 25 50
R2 25 75
R3 50 75
R4 50 100

That takes 4 blind levels to completely double the first blinds with very gentle increases. One reason I like to start at 25/50 is it's highly efficient in terms of the number of chips required.

Anyway, great info BG, and I've added some things to my analysis of tournament structure.

Zombie, I'm the same way. I get invited to tournaments that don't provide that info, and I don't go. We only tell the starting stacks and starting blinds in our invitation, but I'll make the info available to anyone who asks. It takes a lot of room in an invitation to give that out.
 
Last edited:

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