TAG v. LAG style of play in MSSH (Cross-Post from SSSH Forum)

[StellarWind]

Just noticed this X-post. I've said many things in the SSSH thread and some have been misquoted or misunderstood. Here are some key things I said. I use the term TAG to refer to VPIP of 25% or less.

[ QUOTE ]
I'm looking at my 20/40 6-max database for a second-tier site. It includes 172K hands averaging 5.16 players/hand played over a few weeks. I regularly play in this game so my numbers-view is supplemented by personal experience of how things work. The numbers are telling me things I already know because I see them in action everyday.

[/ QUOTE ]
[ QUOTE ]
VPIP doesn't matter nearly as much as most people think. Many successful players have VPIP well above 40%. The ideal range seems to be 28-38%. Players below 25% don't do well at all and are largely extinct in the population of regular winning players.

[/ QUOTE ]
[ QUOTE ]
PFR is more important than VPIP. Low PFR players lose almost without exception. Even more crucial is the difference between VPIP and PFR (calling percentage) which needs to be kept reasonably small. A tight player (VPIP 25-35%) should be calling maybe 7-10% of the time, mostly due to blind defense.

[/ QUOTE ]
[ QUOTE ]
24 players have at least 7000 hands in the database. These are the players who show up everyday and often multitable. 21 of these players are in the green! That's right, 21 winners and only 3 losers averaging +1.48 BB/100.


[/ QUOTE ]
[ QUOTE ]
Only four players (of the 24) have VPIPs under 30% of which the lowest was 26%. The average stats for the whole group are 36/19 and only six players have PFRs of 16% or lower.

[/ QUOTE ]
[ QUOTE ]
Do you see the picture? I can't even ask the question "Do TAGs win in this game?" because TAGs are virtually extinct. Players using the accepted best style have been so badly outcompeted that they are no longer part of the game

[/ QUOTE ]
Now for something new. 198 players in the DB have at least 1000 hands. Let's look at their group results:

PFR <= 12%: 42 players averaging -2.21 BB/100 including 9 winners and 33 losers.

PFR > 12, VPIP <= 25 (TAG): 5 players averaging -1.20 BB/100 including 2 winners and 3 losers. One of the two winners is a talented player at 24.4% or this would be even worse.

PFR > 15, VPIP 25-30%: 14 players averaging +0.68 BB/100 including 9 winners and 5 losers. Only two of these fourteen players are below 27%.

PFR > 15, VPIP 30-35% (LAG-TAG): 19 players averaging +1.22 BB/100 including 11 winners and 8 losers.

PFR > 15, VPIP 35-40%: 25 players averaging +1.46 BB/100 including 16 winners and 9 losers.

PFR > 15, VPIP 40-45%: 22 players averaging +1.22 BB/100 including 11 winners and 11 losers. This group includes several high-volume players whom I consider to be very dangerous postflop. Don't tell me none of them win because I don't believe you. Many of the other players in this group are quite bad.

PFR > 15, VPIP 45-50%: 16 players averaging -0.60 BB/100 including 6 winners and 10 losers. Aren't these players supposed to be burning in Hell!? -0.60 BB/100 is the penalty for being an aggressive clown preflop and usually bad postflop?

What do you think these numbers mean? At what point are you trying to rationalize the numbers to fit your preconceived notions of how poker works?

I can't emphasize enough that there is no "stupid person" filter in PT. Many of these players especially above 40% VPIP are idiots who make asinine coldcalls and play terribly postflop. What do you think these stats would look like if I only included players with postflop skills in the top 50% for this sample?

[disjunction]

FWIW I ran my own personal numbers against these groups in my 200K hands at Party. I had 13544 hands where I won or lost money to TAGs (VPIP 20-30, PFR 16-24, note that these are hands that we specifically clashed, not just sat at the same table, but these include blind steals) and lost 7 bets total. I'll call that breakeven. I had 6706 hands where I won or lost to LAGs (VPIP 40-50, PFR > 26) and lost 9 bets. As a control group I checked out fish (VPIP > 50, PFR < 10), I had 81984 clashes and won 3576 bets. I am about 29/19, so from my perspective I guess these two groups look the same.

More pertinent to the discussion would be how these two groups fare against each other. I am compiling those numbers now (hands from my DB involving a TAG vs an LAG but not me) but I can already tell the sample size will be way small.

[disjunction]

Yeah my Party database was too small, and since all hands in that database have me at the table, once I get rid of hands that involve me I have no hands to work with. I would really need to do this on a mined database to be worthwhile. But since I mentioned it...

In 200K hands, the results I got were:

TAG versus LAG clashes: 450, TAGS win 22.12 bets total

More interestingly:

TAG versus fish clashes: 5587, FISH win 61.41 bets
LAG versus fish clashes: 8495, LAGs win 208

This was at Party 10/20 and 15/30. It's too small of a sample size to conclude anything at all, but it does lend a little support to the "dumb TAG" theory.

[danzasmack]

Just throwing it out there - a 25/15 player that runs hot for some part of 1k hands can show up over 1k hands as 40/30 or whatever.

I had a bunch of statements like that, but they really are all summed up as: being a 40/30 or any specific VPIP/PFR means nothing if you are doing it incorrectly. I think all this post says is to be successful in mid/high sh lhe you have to be able to push tiny edges and that player tight will only get you run over.

[surfdoc]

So, what does this all mean. What is your theory?

It seems that what is happening then is that we are making a series of preflop decisions that are individually losers yet somehow it adds up in the metagame and postflop and makes us winners. Like 3 betting a TAG with QJ. It just doesn't make sense logically that it can be correct.

Similarly, how can this group of guys (the MHUSH regulars) who are fairly dedicated and intelligent not be able to figure out the correct strategy after playing all these hands and studying for so long? Is this the blind leading the blind? Are there forums out there where Brocathmel and friends are laughing their butts off playing 40/30, calling raises from the BB with 47o, and becoming millionaires?

These threads (the "bad TAG" threads) have a very big psychological impact on me, testing the very core of what we are trying to accomplish around here. I hope we get some more responses and perhaps shed some light on this.

[surfdoc]

Is there anyone out there who can post their results showing a win rate over 1 BB/100 at 10/20 or higher playing a VPIP>40, pfr>20 for say 75K hands or more? I am patiently waiting.

[veganmav]

no, but i can show you a winrate of about 1.1bb/100 over 300khands with stats around 28 19 sorted for 5 to 6 handed

[Schneids]

[ QUOTE ]
calling raises from the BB with 47o

[/ QUOTE ]

where's the raise from? how many people are in? [img]/images/graemlins/smile.gif[/img]

[Hock_]

[ QUOTE ]
So, what does this all mean. What is your theory?

It seems that what is happening then is that we are making a series of preflop decisions that are individually losers yet somehow it adds up in the metagame and postflop and makes us winners. Like 3 betting a TAG with QJ. It just doesn't make sense logically that it can be correct.

Similarly, how can this group of guys (the MHUSH regulars) who are fairly dedicated and intelligent not be able to figure out the correct strategy after playing all these hands and studying for so long? Is this the blind leading the blind? Are there forums out there where Brocathmel and friends are laughing their butts off playing 40/30, calling raises from the BB with 47o, and becoming millionaires?

These threads (the "bad TAG" threads) have a very big psychological impact on me, testing the very core of what we are trying to accomplish around here. I hope we get some more responses and perhaps shed some light on this.

[/ QUOTE ]

You're pretty much right a lot of the way. Let me put it like this: (1) I rarely go more than a couple of days without a nitty 100/200 TAG grinder telling me that I'm an idiot fish; (2) See my posts in BBV over the last few months (pretty graphs with nice upward slopes to the tune of $$$$).

If everyone plays the same then it's hard to win, other than those increasingly rare occasions when an uber-fish happens to feel like dropping $10k playing over his head. IMHO, the key to being really successful is understanding how to play solid, but deviating just enough and in just the right spots against just the right players to achieve the metagame results that others have referred to.

Of course, some days I feel like I've found that magic formula and some days I feel like I really am that fish that some players think I am.

[Schneids]

[ QUOTE ]
Of course, some days I feel like I've found that magic formula and some days I feel like I really am that fish that some players think I am.

[/ QUOTE ]

I've said the same thing to others many times the last several months. Very true statement I think of "expert" poker.