Rethinking PokerEV

[ChrisV]

So in like every thread where someone posts a pokerev graph, I see like 5 people confidently assert that a green line being below the red and blue lines indicates that the person is giving up too easily in non showdown pots. I've tried searching, but I have never seen this assertion substantiated. It appears to be based on looking at the graphs of certain strong MSNL players. However what is good in one set of game conditions isn't necessarily good in another, and while it might be generally true that a green line up near the other two represents good playing, theoretically this can't always be the case. For instance, if none of your opponents ever fold postflop, the green line will be below the red and blue lines by the amount of the blinds you fold and there is nothing you can do to change that. There are plenty of other examples, but you should get the point. This is my last 75K hands or so (2/4 through 5/10):



Yes, I'm aware that I run hotter than Jesus. I'm just blessed like that.

What's the quickest way for me to fix this discrepancy? Well, I could start bluffing. Let's say I make pot size river bluffs and pick my spots badly, so that I only win one time in four. What's the effect of this? My total winnings line will drop, but my showdown winnings line will drop dramatically, because every unsuccessful bluff I make carves a big chunk out of my showdown winnings, while successful bluffs invisibly get added to my non showdown winnings. If I post my new graph, 2+2ers will applaud me for bringing the green line closer to the others, whereas in fact I'm making less money.

Any problems with my thinking here? Notice that I'm not claiming that correct strategy isn't in fact to have the lines close together. I'm saying nobody has actually demonstrated that this is so.

[Jamougha]

In a game-theoretic perfect strategy your bluffs typically should be 0-EV. In that case showdown winning should equal total winnings as far as I can see (edit: when playing against game-theoretically perfect opponents [img]/images/graemlins/smile.gif[/img] ). In adjusting to certain game conditions you will probably deviate from that so w/e.

[KeanuReaver]

lower green line means, if anything, you're not bluffing as much as you should

or rather, you're just plain not bluffing much. as a general rule that's a leak.

[Trivial]

do you play on some magic site where you can constantly get paid off on your made hands without needing to balance your lines with bluffs? that is the only circumstance in which i can see your comment being valid

[tozzy]

pokerEV is highly overrated and nowhere near being a proper analysis-tool.

[ata]

If you're value betting too thin in general then your green line (non-showdown) will rise while your blue line will drop (showdown) - which according to your stated popular belief is good, but it actually isn't.

On the flip-side, if you aren't value betting enough, your green line drops while your blue line rises.

The trick is to find the happy medium. You can't objectively determine anything from the lines.

[jalexand42]

[ QUOTE ]
pokerEV is highly overrated and nowhere near being a proper analysis-tool. Can we please stop posting about it now?

[/ QUOTE ]

[ImsaKidd]

if you were playing on a super fishy table you would certainly lose $$ w/o sd because your showdown winnings would compensate.

look up kotkis's graph, its beautiful

[br.bm]

what about your graph in pots smaller than 30BigBlinds?

[schwza]

[ QUOTE ]
In a game-theoretic perfect strategy your bluffs typically should be 0-EV. In that case showdown winning should equal total winnings as far as I can see (edit: when playing against game-theoretically perfect opponents [img]/images/graemlins/smile.gif[/img] ). In adjusting to certain game conditions you will probably deviate from that so w/e.

[/ QUOTE ]

that's not true, even against perfect game-theoretic opponents. your marginal bluff will be 0EV (or the thinnest bluff you make will be 0EV), but some bluffs will be +EV. (i'm pretty sure, not positive).