Q&A #19: Playing a Short Stack Strategy at $5-$10 and Higher

I like your blog Ed but this post really pissed me off. Short stack play is dumb and don’t tell people that it’s not. I know you don’t play taht way. I wrote something at pocketfives about your post here if you want to go see it its at

jared

Hey, jared… Take it easy, bro. Wow, is it that serious?! It’s gonna be okay, man. LOL.

And, just for the record, what I got from Mr. Miller’s post is a thoughtful analysis of the benefits and drawbacks of short stack strategy. I did not understand his post to be an unqualified endorsement of short stack play.

Specifically, Mr. Miller begins his post with

It’s designed to help you get your feet wet without losing too much money - learning an important no limit lesson about stack sizes in the process.

As far as a long-term strategy profitable at high levels… well, it’s not really that at all. It’s too simplistic.

I am currently reading Getting Started in Hold ‘Em and it seems to me that Mr. Miller puts forth sound poker ideas. So, just chill, baby!

Jared lose the ego man. Poker is about winning money. That said, I still think the short stack strategy is highly overated. A winning strategy but for a very low winrate. However, I don’t play in the high stake games. The short stack strategy might be best employed there because of the tough players at those limits.

“100 bb buy in:
Winrate: X PTBB/100
StDev: 40 PTBB/100
Bankroll: 800*ln®/X

20 bb buy in:
Winrate: Y PTBB/100
StDev: 20 PTBB/100
Bankroll: 200*ln®/Y

The whole point of playing shortstacked is that it gives higher risk-adjusted returns.

With $7,000, a 5 PTBB/100 winrate, and a .005 risk of ruin, a deepstacked player will be forced to play at $3/$6 by bankroll constraints. A shortstacked player can play $10/20 NL. At those limits, the deepstacked player would earn $60 per 100 hands and the shortstacked player would earn $200 per 100 hands.

Ponder this:
You have a $100 buy in. You sit in a game against 8 other players who are worse than you. What blinds optimize your absolute winrate? With blinds of $0.01/$0.02, you’ll be deepstacked but earning very little. With blinds of $3/6, you’ll be earning significantly more, despite being shortstacked.

Conclusion:
If your deepstacked winrate (measured in PTBB/100) is N times higher than your shortstacked winrate (where N is the ratio of your deepstacked standard deviation to your shortstacked standard deviation all squared), then play deepstacked. Otherwise, play shortstacked.

Considering that a winrate of 5 PTBB/100 is achievable with a shortstack strategy (tested up to $5/$10), you need to be winning more than 20 PTBB/100 deepstacked to justify playing deepstacked.

Additionally, there’s another relevant argument that I’ve omitted from my calculations. Because playing shortstacked is so formulaic, it’s relatively simple to play six tables at once, sextupling your hourly winrate. Playing deepstacked, it’s difficult to play many tables at once, especially at higher stakes.”

Sluggo from FCP

Do you need a full table IE 10 players for this to work?

Ed, Just read your comment about short stacking short-handed. This interests me because the six max tables are IMO potentially much more profitable due to the number of players raising with marginal hands.

You mention that the main adjustment (not surprisingly) is to defend your blinds more. Presumably this means reraising/calling a raise from LP with more hands than suggested in your GSIH system.

I’m just wondering what kind of hands you’re talking about here? Personally I’ve tried defending with any of the GSIH hands mentioned against a steal raise. That is 77+, AT+, KQ, and usually means moving all in. (I buy in for 20x the BB and I play online).

The problem is that this means I’m only defending with about 10% of the hands I’m dealt, which I don’t think is enough.

So I’ve tried opening this up a little by calling with the hands that you mention can normally be played in LP after a few limpers (22+, A2s+, any two suited broadway) and pushing (or checkraising) if I hit any part of the flop, depending on how vulnerable my hand is.

This means I’m defending about 15% of the time, which seems reasonable as it is only slightly less than 1/6 of the time at a six-handed table. Should I add a few more hands to ensure I’m defending more than 1/6 of the time? Any two broadway for example?

This approach certainly deters people from stealing my blinds, but I’ve no idea if I’m on the right track here. Should I just push with any of these hands? Just call with all of them? Defend more liberally? I’d love to hear your thoughts if you can find the time.

Cheers