Optimal rake betsizing

Here are the rake structures for PokerStars and FTP.

Stars


FTP


An interesting thing to note is if you play 6max on Stars and you got some random dude to sit out at your table, I think this means you'll have rake capped at $2. But that wasn't what I had in mind. What I'm more interested in is the break points at which you pay rake.

I'm going to use NL200 as an example since it'll be easier to understand. Consider when you raise to $6; everyone folds but BB calls. The pot is $6 + $1 + $6 = $13. Right before the flop is dealt, the pot gets raked 65c.

If instead we chose to raise to $5.95, the pot would be $5.95 + $1 + $5.95 = $12.90. This time the pot would be raked 60c since the change is free of rake. I'm too tired to think about how much at all this would affect a winrate, so lets go into something more practical.

Lets look at NL50 HU. Say we normally raise 3x to $1.50. If we are called, that means $1.50 + $1.50 = $3.00 pot and 15c rake or $2.85. If we raise to $1.45 and get called, we get $1.45 + $1.45 = $2.90 pot 10c rake or $2.80.

If we cbet and win in both scenarios, we profit the exact same amount, which is $1.35. It's because you are paying 50% more rake in that specific spot by raising to 1.50 than 1.45. Imagine how often that comes up where you raise pre, cbet and win flop. This way you can save 5c every time it fails and you lose the hand.

Anyway, I'm way too tired to think beyond this right now and I just felt like sharing this.

Also, please point out any logical fallacies if I'm totally off!

nm im wrong

im pretty sure if i ever play nl50 hu again im going to raise to 1.45

Gonna be a hard time to adjust, so i guess it's better to use tableninja ? Waht about raised pots and 3bet ?
I can't see myself raise preflop everytime to the amount minus 5 cents by putting it manually on the bet box >_<

so what does this all mean? any tips about how to optimize ? cliffs? lol

So for a stars 2/4 6max player it is a significant advantage to play 5 handed. Could be worth teaming with people to get 1 person sitting out on every table.

On February 17 2011 10:18 Oly wrote: So for a stars 2/4 6max player it is a significant advantage to play 5 handed. Could be worth teaming with people to get 1 person sitting out on every table.

but what about vpps?

On February 17 2011 11:11 dogmeat wrote: Show nested quote + On February 17 2011 10:18 Oly wrote: So for a stars 2/4 6max player it is a significant advantage to play 5 handed. Could be worth teaming with people to get 1 person sitting out on every table. but what about vpps?

I assume that since effective rakeback from vpps<100% then the reduced rakeback is automatically less than the reduced rake, so it's still good.

i'm going to start raising to 1.36667x just to see what happens

chill out, i was obv trolling

but some nitfags from 2p2 actually started raising to 3.05 or something to get more vpps

Ok this interests me a lot, I think that it could have some serious winrate effects if put into place (most at 50nl but still reasonable at the other limits).

So let's take a starting raise from us and a single caller in a HU pot.

If we raise to $1, the total pot is $2. With rake applied, this is now $1.90.
If we raise to $0.99, the total pot is now $1.98. With rake applied, this is now $1.93.

When we win the pot:

In case 1, we put in $1 and we win $0.90.
In case 2, we put in $0.99 and we win $0.94.

When we lose the pot:

In case 1, we put in $1 and we lose that $1.
In case 2, we put in $0.99 and we lose that $0.99.

So overall, every time we win a pot we make an extra $0.04,
and when we lose we save $0.01.

This has a significant effect upon winrate, and depends on how many pots that you win when you play hands.

-----------------------------------------------------------

You can check your own stats and adjust this to see how it affects you... but basically if you check in HEM to find out your won/lost hand percentages for pots that see the flop that you raised in with 2 people playing, you can say that:

Your won hands/100 will be the won hands percentage, lost hands/100 will be the lost hands percentage (or hands played - hands won).

So this has an effect on your winrate as such by taking into account the stakes played:

winrate boost = (0.04*wonhands% + 0.01*losthands%)/BB

where BB is just a big bet for your stakes.


If you put in like an example of 10% win and lose here (i.e. 20 vpip) at 50nl:

you get an additional (0.04*10 + 0.01*10)/$1 = 0.5BB/100

--------------------

Does this make any sense to anyone who can be bothered to go through it all? It seems worth it at 50nl, and probably still worth it at 100 and 200 seen as you hardly have to do anything...

Very interesting, sick how you raise smaller and the pot is bigger .

This is definitely profitable long run if you do it alot, like OVER 9 THOUSAND times.

poker is really getting degenned a bit too much for my taste..

On February 17 2011 12:50 Gigabeef wrote: Ok this interests me a lot, I think that it could have some serious winrate effects if put into place (most at 50nl but still reasonable at the other limits). So let's take a starting raise from us and a single caller in a HU pot. If we raise to $1, the total pot is $2. With rake applied, this is now $1.90. If we raise to $0.99, the total pot is now $1.98. With rake applied, this is now $1.93. When we win the pot: In case 1, we put in $1 and we win $0.90. In case 2, we put in $0.99 and we win $0.94. When we lose the pot: In case 1, we put in $1 and we lose that $1. In case 2, we put in $0.99 and we lose that $0.99. So overall, every time we win a pot we make an extra $0.04, and when we lose we save $0.01. This has a significant effect upon winrate, and depends on how many pots that you win when you play hands. ----------------------------------------------------------- You can check your own stats and adjust this to see how it affects you... but basically if you check in HEM to find out your won/lost hand percentages for pots that see the flop that you raised in with 2 people playing, you can say that: Your won hands/100 will be the won hands percentage, lost hands/100 will be the lost hands percentage (or hands played - hands won). So this has an effect on your winrate as such by taking into account the stakes played: winrate boost = (0.04*wonhands% + 0.01*losthands%)/BB where BB is just a big bet for your stakes. If you put in like an example of 10% win and lose here (i.e. 20 vpip) at 50nl: you get an additional (0.04*10 + 0.01*10)/$1 = 0.5BB/100 -------------------- Does this make any sense to anyone who can be bothered to go through it all? It seems worth it at 50nl, and probably still worth it at 100 and 200 seen as you hardly have to do anything...

Minraise is 1$ your argument is invalid.

damn, i thought that rake didn't matter that much. but this thread has proved otherwise.

a good read indeed.

tbh spending this much time thinking about rake is probably not as +EV as thinking about your game.

I think the key thing to remember here is when we c-bet, that bet is included in the pot size. So if we raise to .99 and the pot is 1.98 on the flop if we bet more than 1.02 to cbet then we're at a total $3 pot, which is essentially the same as raising to $1 preflop and cbetting anything less than 1.99.

There are definitely situations where this is going to be super helpful though.

lol there should be training sites focusing on rakeback, rake and rewards system abuse and optimal sizing in tourneys , cash etc

On February 17 2011 23:00 Venrae wrote: I think the key thing to remember here is when we c-bet, that bet is included in the pot size. So if we raise to .99 and the pot is 1.98 on the flop if we bet more than 1.02 to cbet then we're at a total $3 pot, which is essentially the same as raising to $1 preflop and cbetting anything less than 1.99. There are definitely situations where this is going to be super helpful though.

What? It shows that temporarily but I'm pretty sure we don't get raked for that. Otherwise that would mean people get raked the full $3 when they open jam flops and everyone folds?

On February 17 2011 19:38 KidPokeher wrote: Minraise is 1$ your argument is invalid.

Weeeel it doesnt make any difference what the absolute raise size is ( it could be $1.49) so it actually doesn't matter.

On February 18 2011 02:59 edzwoo wrote: What? It shows that temporarily but I'm pretty sure we don't get raked for that. Otherwise that would mean people get raked the full $3 when they open jam flops and everyone folds?

Yeah this is correct, unmatched bets don't get raked. Regardless, so long as your preflop sizing is right, postflop you can just then bet full dollars because you can only save one increment anyway.

On February 17 2011 21:48 Zalfor wrote: tbh spending this much time thinking about rake is probably not as +EV as thinking about your game.

Well it only look like 20 mins to calculate and type it up so its not that bad! But I still don't think that it's too bad if you can increase your winrate up to 0.5BB/100 just by taking away a cent from your pfr ^^

good thread imo. Would be cool for table ninja to take this into consideration