18 The Newbie Chronicles: JTG51 vs. Jones

[mdlm]

In a note posted in the Books and Software Forum on 12/10/02, JTG51 wrote: “’Winning Low Limit Hold ‘Em’ (WLLH) by Lee Jones, 2nd edition, was the first poker book I read…For some reason I decided to flip through WLLH the other night. I was shocked at how bad the preflop advice looks to me now….I think the LP advice is even worse. The summary is on page 47. He says with any amount of limpers in front of you to raise with 88, KQ, AT, and A9o. A9o!”

In response to this post, Dynasty said: “I’ve skimmed through that section of the book and it’s atrocious. It may be the worst pre-flop advice in any well known book.” BB King says that the preflop advice given in WLLH is “real bad” for a beginning player. Mason says that he considers WLLH to be “a marginal work.”

I want to see if I can figure out why raising with A9 in LP after any number of callers is bad advice.

I’ll start by making lots of assumptions. First, I’ll assume that Hero has A9 on the button and that SB will call Hero's raise with top 50% of hands and that BB will call Hero's raise with any hand. Second, there are two callers before Hero who limp in with top 25% of hands and always call Hero’s raise. So five players see the flop (two limpers, Hero, SB, and BB).

WLLH assumes nine player tables. According to Acespade, here are the top 25% hands for a nine player game (listed from best to worst): AA, KK, QQ, JJ, AKs, AQs, TT, AK, AJs, KQs, AQ, KJs, QJs, ATs, 99, KQ, AJ, KTs, QTs, JTs, KJ, A9s, QJ, 88, AT, K9s, T9s, 77, Q9s, A8s, J9s, A5s, A7s, KT, A4s, JT, A6s, QT, 66, A3s, A2s, 55.

The first thing to notice is that A9 is not on this list. According to Acespade, A9 is ranked #52 in a nine-handed game.

This means that the two limpers in front of Hero have better hands than Hero 100% of the time. So what Hero is fighting for then is whatever money SB and BB put in the pot.

But SB is calling with top 50% of hands so Hero is even behind SB on average (average SB hand is ranked 42 while A9 is 52). So, on average, Hero has the fourth best hand! Definitely seems to be an error to call, much less raise. In any case, this just leaves BB’s two small bets to fight over. Is this enough and should Hero be raising?

Here is a useful simulation: I put one limper on JdTd and the other limper on KhJs (these are the hands ranked 20 and 21 and since the limpers are calling with the top 42 hands these are average hands), Hero on Ad9s, SB on 5c5s (that is the hand that is ranked 42 and since SB is calling top 84 hands that is an average hand for him), and BB on 9h5d (the hand ranked 88 and since BB is calling all 169 hands that is an average hand for him).

In a hot-and-cold simulation Hero wins 24% of the time, JdTd limper wins 23.5% of the time, KhJs limper wins 27% of the time, 5c5s wins 20.5% of the time, and 9h5d wins 5% of the time.

So there are 10 small bets in the pot after the flop and Hero is winning 2.4 small bets on average and he has put in only 2 small bets. So this simulation suggests that Hero is correct to play A9 here.

Note that all of the players here are profitable except for the BB, so this play entirely depends on the BB being so loose that he calls 100% of all LP raises.

However, I think that this simulation is quite favorable for Hero for two reasons. First, he has the high card. Second, the other four players are holding an above average number of each other’s outs (both limpers have a J, both SB and BB have a 5). I put together this simulation in this way exactly so that Hero would get the benefit of the doubt.

Let’s look at a slightly different simulation: The limpers have QhJd (hand 23) and AcTh (hand 25), the SB has Kc8d (hand 43), and the BB has 8h4c (hand 97). As before, Hero has Ad9s.

Now Hero wins 15%, QJ wins 30.5%, AT wins 21%, K8 wins 22%, and BB wins 11%. So now Hero is wrong to play this hand. Since this is closer to the typical hand, I give this more weight.

Another way to look at this is that approximately 1/3 of the limper’s hands dominate A9. So approximately 55% of the time (1 – 2/3*2/3) Hero is going to be dominated by the limpers. If an A comes on the flop he may lose a lot of money since a weaker kicker will probably fold while a better kicker will win and may raise.

In short, raising with A9 in LP after many limpers seems to be quite dubious. I look forward to your comments.


==>
Comments on Comments

Homer J. Simpson, Angelo Alba, and others point out that the PokerPages tournament rating formula is very different from the one used in real money tournaments. I agree. But my goal here is simply to figure out how to beat an easy game and I have not been able to do that. If I can’t beat play money, I certainly won’t be able to beat real money.

Munga30 suggests scraping PokerPages as a goal and replacing it with something else. My original thought in having this goal was to find the best play money human opposition and beat it soundly. If IRC Poker had been up, I would’ve used it instead. I am open to suggestions. Maybe enrolling in the PokerPages school and playing ring limit games would be a better goal?

Metaname provided a very interesting strategy for playing PokerPages tournaments. I tried it once and I finished #10 out of 109. I spent a lot of time thinking about metaname’s strategy and have quite a few questions.

Metaname writes: “Don’t play any hands at all while the blinds are small (first 40-60 minutes) – muck AA preflop.” In a follow up post metaname says that he tried the system and folded everything for the first 50 minutes so he split the 40-60 range right down the middle. Note that there are only blinds for the first 40 minutes and then after that there are antes as well as blinds. The antes typically add up to 1-2 times the big blind. Metaname, would you suggest playing after 40 minutes?

Metaname also says: “Once the blinds get big and the antes kick in, only play hands in mid-late position.” Are you saying that I should muck AA UTG? And: “only bet if you are the first to enter the pot”. Again, you would muck AA in LP after a few callers? Most critically, metaname says “Raise the pot, fold if reraised”. When you say “raise the pot” does that mean put in a pot-sized raise? If you had AA would you fold if re-raised?

On the issue of river bets, Pudley4 writes “If you make a bet that you know will be called, you don't have to include the "bluff" EV - it is $0.” That’s certainly correct, but that is not what Jones is talking about. He specifically says that your opponent may fold: “Of course, if you think you can make some better hand fold, it may be to your advantage to bet.” He then goes on to say this is “rarely the case” in LL hold ‘em and then proceeds to ignore this branch of the EV calculation. Definitely an error.


==>
Goal Update

This past week, I spent approximately 17 hours on poker: 6 hours reading 2+2, 5 hours studying metaname’s PokerPages system, 2.25 hours playing PokerPages tournaments, and 3.75 hours playing ring games.

I spent $6.57 this week on Zadeh’s Winning Poker Systems. I have spent a total of $445.03 out of my $1000 budget.

An update on each of the four goals (which are to be accomplished by 3/30/03):

1. Read and study Jones’ “Winning Low Limit Hold ’Em”
I have confirmed two out of the three points I need to achieve this goal. A third point (river bets) is pending a discussion with Pudley4. A fourth point (flush draw value bet) is pending an analysis of 10,000 hands.

2. Beat Acespade
Goal Completed on 11/5/02.
Over a period of 100 hours (3600 hands) I beat Acespade’s best lineup at the rate of over 4 BB/hr.

3. Beat Masque World Series of Poker
Goal Completed on 11/17/02

4. PokerPages 85% rating in one calendar month playing 20 tournaments
My PokerPages rating is now 77.07%. I played two tournaments this past week and finished #56 out of 56 and #10 out of 109. This second tournament, which is my third best result of the month, was played using metaname’s PokerPages system.

[Munga30]

I think the UAlberta pokibot game is the best available option. You noted in different thread that it's tough to parse success and failure between "bad" humans and "good" bots. However, I think the diversity of play is one strength of the site.

In a real money ring game, you beat the game when opponents make mistakes against you. Identifying the players that are more likely to make those mistakes and putting yourself in position to profit from them are extremely important game skills. The poki site allows you to practice the identifying part by observiation that may be loosely correlated with the always available statistics. For example, I've seen players with impressive earnings playing very loose and aggressively, but can see from their flop%, graph, and number of hands played that the long run will catch up with them soon. There are are also some very good, friendly human players who beat the game handily and from whom you can learn.

Generally, the poki's play a decent game, but they do not all play the same and all are exploitable. If you play in the expert room with all bots, expect a tight aggressive game in which you can liberally steal blinds. In the regular room, their reasonable preflop standards make it a much different game than what I think is a typical play money game, even with several loose humans. Identifying the differences between the poki's (over time) and beating the overall game will further your overall goal at zero monetary cost.

So, on to an arbitrary goal. I think if you can beat the game for at least 0.1 sb/hand after 5000 hands, you should move on to low limit real money games. With 60 hands per hour (many more possible without humans), that's roughly 85 hours. Not as statistically significant as some would like, but I think it's a good start given your deadline in a month or two. Perhaps John Cole, who's one of the human players mentioned above, could give you a second opinion.

I play under the name "patatty" so please say hi if you get a chance.

[tewall]

A couple of comments. First of all, your reasoning skills are very good. Keep up the good work!

You mentioned that Hero's hand, A9, would be beaten 100% of the time because it's ranked #52 and they're hands would be ranked higher. That's not the way things work. You should be thinking in terms of pot equity rather than hand domination in LL games where many players are playing. The question is whether a given hand in a given position with a given number of opponents has enough equity to be worth playing. You were essentially doing this analysis with the sims you did. One thing I didn't notice you mention is that A9 has position, so if he has a 25% chance of winning a 4 way pot, for example, then he should make money on the hand since he has the best position.

Regarding the tournament advice, I can't imagine a situation where folding AA pre-flop would be correct early in a tournament.

[angelo alba]

Allow me to answer your question with a question of my own, which I hope will inspire the forum members to reply in droves [img]/forums/images/icons/wink.gif[/img] Esp when considering the idea of moving up to higher limits. It is this : How do you beat Lee Jones?

Allow me to explain. You sit down at a $1-$2 game. Everybody's got a thousand bucks and all the other players are willing to stay for days if need be until they triple their grand or go bust. To minimize tells, let's make this an online game.

Now all your opponents are newbies who have amzaing discipline despite their inexperience. Most important they are devotees of Jones. Not only can they quote every sentence by heart, they always follow his advice.

Scenario one: 9 players
Scenario two : 5 players
Scenario three: 9 players, but only 7 are Jonesites, the other two are "Typical low limit players" (whatever that means to you)

So then, how do you beat 'Lee Jones' ??? [img]/forums/images/icons/grin.gif[/img] [img]/forums/images/icons/confused.gif[/img] [img]/forums/images/icons/ooo.gif[/img]

[pudley4]

</font><blockquote><font class="small">In reply to:</font><hr />
On the issue of river bets, Pudley4 writes “If you make a bet that you know will be called, you don't have to include the "bluff" EV - it is $0.” That’s certainly correct, but that is not what Jones is talking about. He specifically says that your opponent may fold: “Of course, if you think you can make some better hand fold, it may be to your advantage to bet.” He then goes on to say this is “rarely the case” in LL hold ‘em and then proceeds to ignore this branch of the EV calculation. Definitely an error.

[/ QUOTE ]

I give up.

I've pointed out numerous times that Jones specifically states "Note that we are not talking about bluffing here". He then goes into an aside about the definition of bluffing, and how in low limit games it is rarely correct to bluff, especially with a good hand, since your opponents aren't laying down better hands.

[pudley4]

</font><blockquote><font class="small">In reply to:</font><hr />
I want to see if I can figure out why raising with A9 in LP after any number of callers is bad advice

I’ll start by making lots of assumptions.

1-First, I’ll assume that Hero has A9 on the button

[/ QUOTE ]

ok

</font><blockquote><font class="small">In reply to:</font><hr />
2-Second, there are two callers before Hero who limp in with top 25% of hands

[/ QUOTE ]

Bad assumption. In LL games, they will limp with anything. The rest of the analysis is based on these limping hands, so I would consider it flawed.

While I agree that Jones' advice isn't the best, I think he makes it easy for a beginning player to get into the game and start winning.

As for playing A9 on the button, I'd raise with zero or one limper, maybe limp with 2 or 3 limpers, and probably fold with more (especially if they are passive players and will limp with many aces.)

[pudley4]

</font><blockquote><font class="small">In reply to:</font><hr />
Maybe enrolling in the PokerPages school and playing ring limit games would be a better goal?


[/ QUOTE ]

I'm not in the school, so I don't know how good the games are. I would bet it's a more appropriate goal than your current PokerPages tourney ranking idea, but I'd still rather see you take the $14.95/mo and put it into a site where you can play .25/.50 games.

[pudley4]

</font><blockquote><font class="small">In reply to:</font><hr />
So then, how do you beat 'Lee Jones' ???

[/ QUOTE ]

Play very aggressively. His advice is very weak-tight, so you can bet/raise your 2nd and 3rd pair, and his followers will consistently fold their middle pairs.

[tewall]

http://www-2.cs.cmu.edu/People/mummert/poker/ discusses errors in Jones' recommendations for hands to open with.

[angelo alba]

tewal: re the site you reccomended, it discusses errors in HEFAP, upon which Jones based his reccomendations. [img]/forums/images/icons/shocked.gif[/img]

Mason wrote an online critique of it here in 2 plus two-- I cannot remember the exact date, so I can't tell you how to access it (perhaps another forum member could) but Mason argued very well and in thorough detail against their methodology and conclusions.

About the ONLY point still in doubt (with me, anyway) is whether KJs is worse than J10s, and, frankly folks, that's not going to alter anyone's game significantly. It won't make headlines.

Pudley 4: Interesting! Carson in his book appears to believe that the games Jones describes are loose-passive, rather than weak-tight( now don't rush me, the difference between loose passive and weak tight is, er... um [img]/forums/images/icons/grin.gif[/img] )
but then Carson accuses HEFAP of the same thing, and apparently feels that no book (His own excepted?) [img]/forums/images/icons/smirk.gif[/img] gives good advice on dealing with very loose and very aggresive games.

( Er.. anyone here play in the California cardrooms who would care to comment?)

Damn! [img]/forums/images/icons/tongue.gif[/img] I dunno, maybe we should we all go back to Poker School? [img]/forums/images/icons/tongue.gif[/img]