ROI after doubling up early

[ymu]

There's an interesting cardplayer article by Matt Matros, To flip or not to flip, where he addresses this issue in MTTs.

Interestingly, he makes the same point as Suzzer derived from the maths on the other thread about this - that the benefit of doubling up early is enough that you should be willing to take a slightly negative edge early on in order to do so.

How much does his argument need to be modified when applied to STTs?

[suzzer99]

FYI - I never said you should take a slightly negative edge. I said the edge you should look for is some function of a bunch of factors. Some of these factors might push your minimum edge lower than one might expect. But I don't see ever willingly taking a 45/55 early.

[ymu]

[ QUOTE ]
FYI - I never said you should take a slightly negative edge. I said the edge you should look for is some function of a bunch of factors. Some of these factors might push your minimum edge lower than one might expect. But I don't see ever willingly taking a 45/55 early.

[/ QUOTE ]

This is from one of your posts on the other thread:
[ QUOTE ]
Interestingly, if those #s did hold up it would suggest taking even slightly negative coinflips early, from an ROI pov. Not to mention from an hourly rate pov (double up or start a new one).

[/ QUOTE ]

Not to suggest that you'd want to be as negative as 45/55, but your comment on that thread reminded me of the Matt Matros article, hence my post. Matros suggests that around 48/52 is worth taking early on in an MTT. I was just wondering whether anyone thinks his argument is worth considering for an STT - it seems to be pretty relevant to the discussion on these two threads.

[suzzer99]

I think when I wrote that I was misguided about what 100% roi actually meant. It wasn't occurring to me that busting out was -100% roi. + I was a moron about all things SNG back then. Anyway, like I said now I think ballpark it might be somewhere between 55/45 and 60/40 in a SNG - towards the lower end if you play continuous.

[HesseJam]

mtt is way different from sng. they take wayy longer. u can count on busting after 2 hours without showing for anything. -> huge bearings on hourly rate.

if you double up early in an mtt it could well mean that your roi goes from 100 to 350%. moreover, it could mean that your itm chance increases from 2% to 10%

[futuredoc85]

absolutely not worth considering for stts. i havent read othe posts so forgive me if im repeating. MTTs are all about making the FT and finishing in the top few b/c of the top heavy payouts. STTs are all about finishing ITM. Therefor early on you should be extremely tight and not risk it on a flip, b/c a double up in the first level is not nearly as advantageous in a STT as it is in a MTT. I wouldnt even take a 55-45 at the first level of a stt if i knew for sure thats what it was. Maybe at the $109s and up i would reconsider, but below that no way.

[futuredoc85]

that was in response ymu, not OP

[schwza]

how'd you figure this out? is there a way i can figure it out for me?

[HesseJam]

the coolest way is by learning sql and make a query to your ptrack database.

im uncool and do this through the external data function in excel and some minor footwork

[beeny]

[ QUOTE ]
Latest research yielded lower figures. My ROI once I reach between 2800 and 3200 chips during the first 3 rounds and at least 8 players left is around 110%.

Conclusion: A real coinflip (50:50) at even stakes (like only SB and BB in the hand) yields an expected average ROI of 5% . Thus I should probably not engage in true coinflips. A 55:45 I should definitely do.

[/ QUOTE ]

I'm a little confused. Don't you have to account for the times where you reached 2800-3200 chips by methods other than a 55/45?

Aren't most of the times you get to 2800-3200 chips from AK/AA/KK and not the 55/45 flips? .. or did you already filter those out too?