Thoughts on 'run it twice'?

[superwomble]

I noticed last night that Stars has the option on Zoom tables to 'run it twice' for no extra rake if both players in the hand have the option selected.

I also read some advice from Doug Polk who said that if you ever have the option to run it twice (for no additional rake) you should do so. The reason he gave is that you should look to limit your varaince as much as possible and RIT does that.

What are your thoughts on this? My first thought was that I can't see why this is - surely most of the time when you get it all in you do so believing you have the best hand, so why give your opponent two chances to suck out on you? I can see the positives for when you get it all in with a good draw, to give yourself two chances to hit it, but why do it every time?

[Doobs]

Getting the same EV with less variance should always be better.

I don't think that always getting it in ahead is optimal.  If you are 40% in a hand and you are getting 2/1 on a call, you should call in cash.  You also should bluff enough, and you are going to be behind those times.

If you are worried about getting sucked out, then surely it is better for half a stack rather than a whole one.  Though if this is effecting your game, you need to strengthen your mental game.  Maybe read DaveShoelace's book?

The first point above is way the most important though.  Craving extra volatility is bad.  You need a smaller bankroll for less volatile games with everything else equal.  You are less likely to go bust/ should go bust slower, with everything else equal.  My box is ticked. 

[StuartHopkin]

why give your opponent two chances to suck out on you?

Think this is where your problem is?

I believe the more times you run it the more likely you are to realise your expected value.

If you could run everything 100 times then everyone's sad face graphs showing all the EV they lost out on wouldnt happen as often?

[arbboy]

I would only run it twice in a game i wasn't 100% comfortable playing in and taking a shot or if everyone in the game was comfortable.  I like the angle of putting people who are taking shots outside of their bankroll/not good enough to play the game skill wise under pressure by only running it once, potentially sending them on tilt, playing badly because they lost a key pot.  I think there is a lot of long term EV to be gained if you see the bigger picture.

Obviously running it twice/three times doesn't change the EV at all as Doobs correctly says only reduces variance.  I think sometimes adding variance to games can make you more profitable longer term.

[Rexas]

I like running it twice, although unfortunately they rake chopped RIT pots at DTD so it's probably not the best thing to do there :p

I just figure if I'm behind I've got two chances to catch up and if I'm ahead I've got two chances to hold, and if people want to get it in lighter against me because I'm happy to run it twice then that has to be a good thing, right?

[SuuPRlim]

Running it twice is fantastic for poker, any pro player should want to run it twice in general and nothing to do with variance. It's fantastic for casinos too, they should be desperately encouraging  people to run nit twice and I'm pretty shocked DTD rake chopped RIT pots as it's very very good for the ecology of cash games. They should defo not do that.

The big misconception regarding run it twice is that it matter in the slightest...theoretically it makes absolutely zero difference at all, it's a totally meaningless thing. I have had hundreds (and i don't think I exaggerate) of conversations with people about it and they just refuse to accept that it is a totally neutral thing to do.

Think of it like this;

If I say "lets flip a coin for who pays for dinner" we're now a straight 50/50 for who is paying for dinner (not if you're vs me obviously, I never lose these )

and you say "let's do best of 3"

What are the odds now?

[Karabiner]

Running it twice is fantastic for poker, any pro player should want to run it twice in general and nothing to do with variance. It's fantastic for casinos too, they should be desperately encouraging  people to run nit twice and I'm pretty shocked DTD rake chopped RIT pots as it's very very good for the ecology of cash games. They should defo not do that.

The big misconception regarding run it twice is that it matter in the slightest...theoretically it makes absolutely zero difference at all, it's a totally meaningless thing. I have had hundreds (and i don't think I exaggerate) of conversations with people about it and they just refuse to accept that it is a totally neutral thing to do.

Think of it like this;

If I say "lets flip a coin for who pays for dinner" we're now a straight 50/50 for who is paying for dinner (not if you're vs me obviously, I never lose these )

and you say "let's do best of 3"

What are the odds now?

They also take full rake from hilo split pots which is one reason why I lost interest in playing there.

[SuuPRlim]

so why give your opponent two chances to suck out on you?

I can see the positives for when you get it all in with a good draw, to give yourself two chances to hit it

I think the aspect of Run It Twice you're kind of forgetting to account for (and this sounds really obvious but it's very common) is that the pot is split into two. So in your first example, as much as you are giving your opponent 2 chances to suck out of you...you're at the same time FORCING him to suck out on you twice if he wants the whole pot.

In the second example, it's the reverse, you're giving yourself 2 chances to hit it...but again if you want the whole pot you gotta hit it twice.

So actually the two examples seem massively different...but in reality they are identical. The maths is like this

Lets take a standard 70/30 Spot (you have AK vs 56 on A47 something like that)

You're 70%, you run it twice. 70% of the time you win the whole thing....30% of the time you lose the whole thing. Simples.

Run it twice. You're now only 49% to win the whole thing (70%*70% - yuo have to win both boards twice) so that sounds bad. HOWEVER, as a result you're now only 9% (30% * 30%) to lose the whole thing.

If the pot = £1000, run it once your equity is worth £700.

Run it twice 49% of the time you get it all (£490) 9% of the time you get none (£0) and 42% of the time you get half (£210 (£1000/2 * 42%) £490+ £210 +£0 = £700


Do it from the reverse, you have the 56 vs AK on A47...

Run it once, 30% of the time you win it all, 70% of the time you lose it all = £1000 pot = £300 equity.

Run it twice, you only win it all 9% of the time, but now you only lose it all 49% of the time.

9% win (£90) 49% lose (£0) and 42% chop (£210) = Equity when running it once £300

You can run it 1 time, 2 times, 5 times, 100 times. The win% and lose % changes but the bottom line EV remains identical.


Problem with all this though is we're humans, it doesn't matter how sensible and level headed you are, everytime you run it twice win the first and lose the second you're always going think "Dammit should have run it once" and you'll be told be everyone at the table too.
Everytime you run it twice, lose the first and win the second you'll think "Thank god i ran it twice" and no doubt someone else will say it to you too
Everytime you run it twice and lose both you'll feel hideously unlucky and every time you run it twice and win both you probably couldn't care less about why you ran it twice

[atdc21]

Say i have quads and you need one card that can make a straight flush, if we go once it is possible i could lose. go twice i can win mostly or never lose, 3 times i can never lose

[atdc21]

be fun talking you into it 

[SuuPRlim]

Say i have quads and you need one card that can make a straight flush, if we go once it is possible i could lose. go twice i can win mostly or never lose, 3 times i can never lose

You telling me if you RIT and he hits it on #2 you wont be tilted?!?!

[Doobs]

Running it twice is fantastic for poker, any pro player should want to run it twice in general and nothing to do with variance. It's fantastic for casinos too, they should be desperately encouraging  people to run nit twice and I'm pretty shocked DTD rake chopped RIT pots as it's very very good for the ecology of cash games. They should defo not do that.

The big misconception regarding run it twice is that it matter in the slightest...theoretically it makes absolutely zero difference at all, it's a totally meaningless thing. I have had hundreds (and i don't think I exaggerate) of conversations with people about it and they just refuse to accept that it is a totally neutral thing to do.

Think of it like this;

If I say "lets flip a coin for who pays for dinner" we're now a straight 50/50 for who is paying for dinner (not if you're vs me obviously, I never lose these )

and you say "let's do best of 3"

What are the odds now?

They refuse to accept it is totally neutral, because it isn't.

Your dinner example is 50/50 you pay for dinner whichever way you go.

Doing two £50 flips rather than one £100 one means you lose your £100 only one time in four rather than one in two.

Sure some would prefer the extra volatility, but that doesn't mean it is better.  

[SuuPRlim]

They refuse to accept it is totally neutral, because it isn't.

By Neutral I mean, there is zero advantage to be gained from any amount of times you run it.  Which I know you agree with.

Lets roll a dice. You can have £1000 and 16-1 on any number one roll....Or you can have 7-1 on 100x £100 rolls - which one you taking?

[Doobs]

They refuse to accept it is totally neutral, because it isn't.

By Neutral I mean, there is zero advantage to be gained from any amount of times you run it.  Which I know you agree with.

Lets roll a dice. You can have £1000 and 16-1 on any number one roll....Or you can have 7-1 on 100x £100 rolls - which one you taking?

Your examples are completely different scenarios to the one in the OP.   

Same EV less volatility is superior to same EV more volatility. 
Different EV different volatility, it depends. 

In your latest example, you of course take the £10k getting 7/1 on a 5/1 chance.  Just way clear of getting 1k at 16/1.

Hope you aren't doing the maths side of your bookmaking business.

[titaniumbean]

They refuse to accept it is totally neutral, because it isn't.

By Neutral I mean, there is zero advantage to be gained from any amount of times you run it.  Which I know you agree with.

Lets roll a dice. You can have £1000 and 16-1 on any number one roll....Or you can have 7-1 on 100x £100 rolls - which one you taking?

Your examples are completely different scenarios to the one in the OP.   

Same EV less volatility is superior to same EV more volatility. 
Different EV different volatility, it depends. 

In your latest example, you of course take the £10k getting 7/1 on a 5/1 chance.  Just way clear of getting 1k at 16/1.

Hope you aren't doing the maths side of your bookmaking business.

zing