Is it just me or is the whole "Determine Villain's Open Raising Range" formula completely and utterly pointless?

First of all it's not valuable or good for a better understanding of NLHE at all, mostly becuase it's incorrect, villain will change his openingrange depending on mood, tableconditions, precense of ratolher etc.

And the only difference between this and just taking the pfr percentage directly is that he will have somwhat lower pfr then openinraise because people will have opened in front some times and he will not 3bet all hands he would have opened with. So in that sense this formula will be more reliable, but that really doesn't mather at all since you made approximations about openraises in front and his 3-betting percentage anyway. And the formula doesnt take into account how the player responds to limps in front of him, he might tighten up or loosening up his raising percentage depending on that too.

And last but not least, you used his overall pfr, not the pfr in the UTG+1 seat. There's probably a significant difference between those, even bigger if he would have been in the cut-off or the button. This was probably just a misstake by your part but it changes the result dramaticly.

The fact that you used the overall pfr and not his pfr from utg+1, and the fact that you used approximations about the openingpercentages in front of him aswell as his 3-bet percentage makes the result vastly incorrect. You could aswell just filter PT for the amount of players at the table and look at his pfr from utg+1 and just add a percent or two and get a more reliable result. But that's not really the big problem with the formula. The big problem is that his openingrange range will not be consistent, it will change depending on many factors. Using PT stats and formulas for this is not the way to go. If you know the player and now how he plays, what mood he's in all the other variables you get a better result just using logic and put him on a range you think he's opening with and gow with that instead of that pointless formula.

This got quite long and sounded more harsh then I intended it to do. I haven't really followed your series but I like what I have seen (except said formula) and I actually think I will watch it all when I have time. You're doing a great job and I know that many pokerplayers out there really should take a look at your series for their own best. But I just reacted on that you spent the first 20 minutes on this video on such a pontless formula.

This does not make the formula incorrect. It's a general formula for a general figure, just like PFR% is a general figure in the first place.

True, I might have used the word pointless" a little recklessly,, sry about that. But the truth is that you most likely get a much better number by looking at the opponent, his tendencies, his flow atm and the table and just use common sense to come up with a range. And use that range to determine if it's better to repop of flat AK.

Everything the forumla does is take into account that the pfr% from one possition will not be the same as the openingraise% from that position because of the fact that people can open in front, it doesn't even take limpers into account (even though it will make very little difference, but approximating pfr with openingraise from utg+1 or using this formula will make very little difference to). And spending 20 minutes or 1/4 of the video on it (and missapplying it) is, imo, pointless.

I agree I probably shouldn't have spent as much time as i did on the formula, but i don't think it's pointless. Once I got into explaining each part, I wanted to make sure it was fully understandable (at the cost of time/redundancy). A better approach might have been saving everyone's time and just explaining once through what each value of the formula represents and letting people plug-n-play. However in terms of the usefulness of the formula, like many of the poker math topics, the formula takes some estimation and guess work to plug in values. I don't think that makes it flawed, at least in comparison to other math topics like plugging in how often you think someone folds into the fold equity calculation... it's an educated guess, just how using the preflop raise% from PT is a (more accurate, imo) educated guess that gets refined through this formula to pull out a range.

You're right though I should have better explained about using his positional PFR%, but I do encourage you to go look at your PT DB and see how much a really tight player's opening range changes by position. I'll give you a hint... most do not(at least at 6max). However for purposes of completeness (and if someone wanted to do this formula for a more laggy player), you're right I should have mentioned using that particular figure.

One reason this series is very difficult to create is because I'm trying to not only make it accessible to a wide range of people (in terms of their Math backgrounds) but I'm also trying to use somewhat real life examples so even if people don't want to sit down with a pen and paper to figure this stuff out, they might learn something about situations they encounter. At least, that's the goal. Whether I succeed or fail in those things is ultimately up to all you guys.

Thank you klb for the feedback though, I do really appreciate hearing what everyone has to say -- good or bad.

Aaron

So full of informations... my head is about to explode !

I think some homework/practique sheet could also help those who want to work on it. Do you think about putting some homework for peoples who wish to ? (I mean like 3-4 Questions with calcul to make on something like word.doc or straightforward in forum text.. and we discuss about the answer week later)

Will do my best to incorporate theses thing in my play, thanks

wow, I didnt expected to take so long time to do a video like this.. (ok this one lasted 1h20min) then I clearly understand you

I was writing down your p calculations for fold equity and something came to me that seemed simpler to me.

In general if we are a 3:1, thet means we have 4 situations (3 negative and 1 positive).

Our 1 postive cancels out 1 of the negative, leaving of course 2 negatives out of the 3 negatives.

So we need him to fold to cancel out those negatives.

So if he folds 2 out of the 3 negatives the 3 negatives are cancelled out.

2/3 is 66%!

I'm a logical rather than mathematical person so this could be full of holes.

But it seems to be interesting at least.

If were a 4:1 dog we need him to fold 75% of the negative times.

If we're even money, 1:1 we don't need him to fold at all.

If he's a dog 1:2, we have to fold for him to win the pot.

Seems logical!

Now this doesn't take into consideration what's in the pot, only that we win it.

So if we only try to win it when we have the correct pot odds we should be ok!

Could you run some math on this idea and see what floats to the top?

May be my flash was a brain cramp.


Thanks!
sudic

min 26 - isn´t our EV only 12.73?

min 26 - isn´t our EV only 12.73?

can you elaborate on why you think this? i just redid the math and it seems right?

WoT

EV=(our equity) * (what we win)- (villain´s equity) * (what we loose)
EV= 55.5% * (40+15) - 44.5% * 40
EV= 0.555 * 55 - 0.445 * 40
EV= 30.525 - 17.8
EV= 12.725

or

EV= (our equity)*(total pot) - cost of our call
EV= 0.555*95 - 40
EV= 52.725 - 40
EV= 12.725

to me seems wrong to count our call in "what we win"

Anyway "Math of NLH" are best poker videos I have ever seen. Thx a lot for them.

SugarNut made me aware of an algebraic error I made in this episode around the 45 minute mark. it doesn't drastically change the result or spirit of the calculation, but it is wrong.

So here's how it is in the slide:

EV(total) = 370x – 12.85(1-x)
EV(total) = 370x – 12.85 + 12.85x
EV(total) = 382.85x – 12.85
12.85 = 382.85x
x = 12.85 / 382.85
x = 3.4%


Here's how it should be:

EV(total) = 370x – 12.85(1-x)
EV(total) = 370x – 12.85 - 12.85x
EV(total) = 357.15x - 12.85
12.85 = 357.15x
12.85 / 357.15 = x

x =~ 3.6%

The error occured on line 2 where i distributed 12.85(1-x) incorrectly. Sorry about that guys.

Also, Luzhin64 above who posted back in march, it looks like he is correct as in the video I somehow added in an additional $40 to my equation. So around the 26 minute mark the incorrect equation is:

EV = [Our Equity] * [what we win] – [Villain’s equity] * [what we lose]
EV(call) = 55.5% * (40 + 40 + 15) – 44.5% * (40)
EV(call) = 52.73 - 17.8
EV(call) = 34.93

and the correct way is is how Luzhin64 wrote it above:

EV=(our equity) * (what we win)- (villain´s equity) * (what we lose)
EV= 55.5% * (40+15) - 44.5% * 40
EV= 0.555 * 55 - 0.445 * 40
EV= 30.525 - 17.8
EV= 12.725

I'm not sure how that extra $40 crept into the equation. Must have been a typo or absentmindedness. How embarassing, sorry guys! At least the faithful DC'ers found my mistake and got it corrected for everyone else. Also at least it didn't hose up the process, just slight errors in calculation... so as long as you've learned the process you'll probably calculate correctly :-)

Thanks
WoT

At 28 minutes you are determining villains' range for a call of our 3bet, deciding that such a range would be TT+ and AK. You go on to say that this accounts for 3.5% of all holdings. What confuses me is that in all previous episodes you have taken pains to point out that we should remove from our opponents range any possibilities that involve either the community cards or our own pockets. Here we have AsKs, so surely we should remove options involving these cards from his range: If we do that then his calling range is more like 2.5%.

This in turn means that we will collect the pot 77.7% of the time, not the 69% stated in the video. Our EV is then $40.65, not $34.89.

Of course I could be the gibbering idiot so I'm a little confused as to what I should be doing. Please enlighten.

Around the 48 minute mark, you list the combinations of hands. If you hold the As, there is only 9 ways to make AKo, AQo, AJo. Also, there are only 12 combinations of KQo regardless.

Thank you a lot WiltOnTilt. I thought I knew pretty much about maths and so on but these videos have been a good eyeopener. I believe that my view of certain opponents acts are now more clear to me. Hopefully I can turn my play from break even to winning.

Yours Madaa

SugarNut made me aware of an algebraic error I made in this episode around the 45 minute mark. it doesn't drastically change the result or spirit of the calculation, but it is wrong.

So here's how it is in the slide:

EV(total) = 370x – 12.85(1-x)
EV(total) = 370x – 12.85 + 12.85x
EV(total) = 382.85x – 12.85
12.85 = 382.85x
x = 12.85 / 382.85
x = 3.4%


Here's how it should be:

EV(total) = 370x – 12.85(1-x)
EV(total) = 370x – 12.85 - 12.85x
EV(total) = 357.15x - 12.85
12.85 = 357.15x
12.85 / 357.15 = x

x =~ 3.6%

The error occured on line 2 where i distributed 12.85(1-x) incorrectly. Sorry about that guys.

You distributed correctly in the video.

The check works out....

0 = 370(12.85 / 382.85) + (-12.85)(1 - (12.85 / 382.85))
0 = 370 * 0.033564059030952070001305994514823 + -12.85 * 0.96643594096904792999869400548518
0 = 12.4187018414522 - 12.4187018414522
0 = 0

if your ev calculation is EV = [EV(fold)] * x + [EV(call)] * (1 - x)
Then distributing -12.85 would yield a calc that looks like this ..
EV = 370x + [(-12.85)(1 - x)]
= 370x + (-12.85 + 12.85x)
= 370x + 12.85x - 12.85
= 382.85x - 12.85
x = 12.85/382.85
x = 0.033564059030952070001305994514823

Someone correct me if I'm wrong.

I can see how SugarNut came to his conclusion somewhat.
Maybe he changed around the origincal Fold Equity equation to...
EV = [EV(fold)] * x - [EV(call)] * (1 - x)
In that case x would be -.0359793.
But even then the answer seems erroneous being negative and all that. But i can see the correlation between 3.6%.

Anyone want to clear that up?

anyone have some input on my math like 2-3 posts up? im uncertain if that is right or not

was arguing with someone on irc they said i needed him to fold like 30% ?!?

im so bad at math... lol