I saw a new stat today used in a video that I thought was interesting. I had to download it from PT4 warehouse. It is Faced 3b percent. Not to insult anyones intelligence but it shows the percent that you opened from a position and faced a 3bet. This got me curious.

Assuming you knew an optimal 3bet percent based off of optimal opening ranges you could determine what the 3b percent is for say HJ vs UTG, then CO vs UTG, then BTN vs UTG and so forth. How do you take that and determine what this stat should optimally be? Would this be something as simple as the arithmetic or geometric mean of all of those 3bet percentages vs a UTG open? Thanks for any help.

I don't think it would quite be a linear relationship.

That's what I was thinking but then thought I might be over complicating my thought process. I am decent with math but not really smart enough in statistics to logically think through it.

Most three bet ranges will typically consist of CO being a subset of HJs 3b range which is a subset of buttons 3bet range vs the same position. I am not sure if we could take buttons threebet range and say that UTG should be threebet that percent of the time from the positions of CO, UTG, and BTN combined then figure out the SB and the BB or if I am way off base with this thought process as well.

I don't think this is particularly useful.

Do you mean my thought process isn't useful or determining the statistic I am trying to determine?

I think it would be useful in enabling you to determine if you are being threebet more or less than might be optimal. It would enable you to determine how aggressive a population is being in this situation or not being in a particular situation.

Both.

The statistic you're looking at seems to be a function of your RFI, and if you'd like to adjust to opponents then looking at how they are deviating vs. you specifically is likely not useful until you are at mid-high stakes playing against other regs consistently.... And even in that situation it's likely easier to visualize their ranges utilizing their 3b statistic as a function of both position and then "vs. hero" and these stats exist in both pt4 and hem.

I'd argue its likely more effective to just use the simple 3b stat as a function of position rather than looking at whatever it is this stat is.

Thanks for the input. I play on Bovada Zone currently. The tables are anonymous so I won’t have data on specific players. I like to look at the population tendencies though. I am still learning the different features of PT4 but you might be right. I’m away from the computer right now but it seems like your saying there are stats (or you could create stats) that are position based for how Often you are getting 3b in a certain position from certain positions. Is that right?

I did some work on this stat last night too and found out it is basically a sum of each positions threebet vs the specific position. So for my UTG open faced 3bet it is how often I was 3b from CO, HJ, BTN etc added up.

It is very simple to create an alias for zone given all hole cards revealed and look at the populations average 3b frequencies in any situation (for example by position and vs. position).

can I suggest one thing:

stick to snowie preflop advisor religiously for a while (2000 hands or so)
esp Blind defense, 3b, fold 3b, 4b

you will see your stuff converging to optimal / close enough..

you need not reinvent the wheel in my opinion..everything is already done and dusted.

Snowie preflop advisor is pretty trash because all the ranges are calibrated for pot-sized 3-bets and 4-bets, which isn't what happens in theory (or even in practice, in my experience).

For instance optimal 3-bet sizing BU vs CO is 75% pot
SB vs BU is 150% pot
BB vs BU is 200% pot
BB vs SB is 125% pot
optimal 4-bet sizing BU vs SB/BB is MIN to 50% pot
I could go on...

what is the open sizing in each case?
source?

There is no source. He's just stating his opinion as a fact.

Source: monkersolver preflop sims. Give it the option to 3-bet less than 200% pot BBvsBU, and it rarely takes it, opting for 150% only sometimes. Highest EV option is to go bigger and more polar. This is versus a 2.5BB BU RFI, but it chooses a similar 3-bet size (in terms of raw # of BBs) versus smaller opens.



(These are taken from Zenithpoker's preflop library btw.)

what's the rationale behind BB vs BU 2pot??
The remaining I sort oF get it

what is the sizing of SB open in BB vs SB 125% pot?

BB doesn't want to 3-bet too linear because that reduces the EV of its calling range (too many medium-strength hands would be missing). It's getting decent odds to call with a wide range, so it just 3-bets the top of its range and adds in bluffs for balance. Because it's so polar, it uses a larger size to reflect that polarity. That way it gets more value with its strong hands, and pushes more fold equity with its bluffs. Hope that explanation helps!

To your second question it's SB opens to 3BBs, BB 3-bets to 10.5 BBs

Are you asking for what the equation needs to be to work out the stat? I cannot find that stat in the warehouse. However, I would imagine if you went into Configure->Statistics, you'd be able to see the statistic's equation breakdown.

My best guess would be the following:
BTN faced 3b percent = (SBvsBTN)+(SBvsBTN*BBvsBTN)
So, let's say SB and BB both 3b the BTN 20% of the time. The above would equate to you facing a 3b 24% of the time; 20%+(20%*20%)=24%

thank you - makes sense - i haven't tried this yet

I have snowie 3betting BB vs SB 3x to 9x - any comments on that

Thank you. I ended up figuring it out. Still learning the ins and outs of PT4 and forgot you can do this.

Thanks. Im using a python script to convert hands for villains by changing their names to alias and run analysis on them. Your recommendation ended up being more useful as I was able to see the exact population frequencies one position is 3betting vs RFI for other positions.

I was trying to get an idea of how much I needed to narrow my continuing range that I am currently using vs three bets. I am using a preflop solution from monker solver and I had a feeling the continuing ranges for vs threebet were too wide given the narrow threebet ranges I was seeing.

Looking at a population's 3bet range and then comparing that to a preflop sim to identify how to play your range can be a beat misleading due to all the postflop leaks that occur from villains--and sometimes from you depending on your skill level. You're very likely to over realize equity vs. the population if you are a decent to good player compared to what a preflop sim might suggest.

This results in a situation in where the preflop solver isn't particularly helpful (and this also applies to other situations in which future streets have high EV exploitative plays). Comparing to a sim is only going to give you a hard baseline.

Consider the example where a villain continues vs a cbet at 100% frequency with only calls, but then folds turn vs. a cbet at 100% frequency. This is difficult to sim, but the optimal play to just bet everything is very obvious. In future streets where villains deviate and more specifically over-fold, this results in a large incentive for Hero to put as many hands in possible to previous streets so they may realize that future opportunity to secure EV. This is similar to the logic used when you are playing vs. a fish and want to try to vpip as much as you can.

Just something to think about when you're utilizing any solver software.

All good points that I should have considered but didn’t. Thank you for that.