w34z3l's Free Database / Stat Analysis Thread

Posted 8 years agoEdited 8 years ago

Hey Guys!

If the thread is not clear, you can post your stats for analysis here. Seeing as it may not always be super obvious what exactly needs posting for a basic stat review, here is a rough guide. You can post here as often as you like and it is completely free.

Forum Rule - You must have 10 good posts or be a valued PokerVIP member to receive this free analysis.

1. Have a Sample

I don't mind looking at small samples, but when it gets to below about 20k you are probably wasting everyone's time and not getting a lot of benefit in return. See if you can have at least 20k hands. If you can have 50k or 100k even better. The sweet spot is usually somewhere between 50k and 150k. (Sample sizes more than this often contain conflicting data where our play style has change, but not always. Assuming no big changes in game plan, the more hands the merrier!)

2. Post stats by position.

Both HM2 and PT4 have the option to view stats broken down by position.Please make sure you include this in your screenshot. It's possible for an overall stat to look fine, yet serious problems are identified when analysed positionally.

3. Feel free to include results graph showing both the red and blue line.

This is not 100% necessary. Actually, what we are saying is that there is basically no point in posting your graph unless the red and blue lines (showdown and non-showdown winnings) are included. If you do post a graph, put it in big-blinds ideally, but feel free to include $ alongside it if you prefer.


Ok, now for the stats we need for some solid basic analysis.

VPIP
PFR
RFI
3bet
Fold-to-3bet (for PT4 make sure this is "fold to 3bet after raise)
cold-call (for PT4 make sure this is "call PF 2bet" not cold-call
Fold to cbet on flop/turn/river (Include for 3bet pots if you think it's a problem)
Cbet on flop/turn/river (Include for 3bet pots if you think it's a problem)
Raise vs Flop Cbet
WTSD
WWSF
WSD
4bet range (not frequency)
Fold to 4bet
River call efficiency
win-rates



The following is not mandatory, but for more advanced analysis you can include.

Bet vs missed cbet on the flop (In PT4 this is called "float flop")
Bet vs missed cbet on the turn (In PT4 this is called "float turn")
Bet vs missed cbet on the river (In PT4 this is called "float river")
Bet vs missed cbet OOP on turn (In PT4 this is called "probe turn")
Bet vs missed cbet OOP on river (In PT4 this is called "probe river")
Delayed Cbet IP
Delayed Cbet OOP
Fold to Delayed Cbet IP
Fold to Delayed CBet OOP
Skip flop/turn/river cbet and check/fold (In PT4 "fold vs flop/turn/river float")
Skip flop/turn/river cbet and check/raise (In PT4 " raise vs flop/turn/river float")
Skip flop/turn/river cbet and check/call (In PT4 "raise vs flop/turn/river float")



Ok, so we are not saying this is everything we need for a complete DB analysis. It's obviously way more complex, and once problems are found we need to run filters and understand those problems in more depth. Anyway this should be a good starting point.

Feel free to disregard any of the above information. It doesn't bother me that much, it's you that will miss out Wink
FG_SpaceLord

Last Post 6 years ago by

FG_SpaceLord

92

Posts

25,618

Views

Copy post URL
https://www.pokervip.com/thread/view?forum=poker-strategy&slug=w34z3ls-free-database-stat-analysis-thread&nav=568a8c09d390435e438b45a7
4
Posted 8 years ago
Great! Very excited to see this thread! Smile

It's perfectly timed for myself

After quite a successful start to my poker journey (Winrate of about 6.5 over 100k hands), the inevitable was going to happen... A bad run... Ups and downs but hardly any profit... My winrate has dropped to 4.3. A painful 70k hand break even stretch!

The weird thing is... I genuinely feel like I'm playing the best poker I've ever played...

Would this be considered a long stretch or is it quite common?

I am mainly playing 5NL Snap after a really bad run at 4NL reg tables. I know the variance is usually higher at Snap tables but they're full of fish!

It should be noted that I moved to 888 from Bwin just as this stretch began. I find the players are a lot poorer on 888... Yet I'm not winning...

Maybe you could take a quick look at the past 70k sample for me?

Cheers Adam!

General

Attached Image


Attached Image


Attached Image


Position

Attached Image


Attached Image

Posted 8 years ago
I can send over my graph with red & blue lines when I get home from work soon! Apologies for not including. Don't have them available on this computer
Posted 8 years ago
Here's the past 70k hand break even stretch with red/blue lines

Attached Image
Posted 8 years ago
Adam,

Loving this idea and of course I am in!

Graph
Attached Image


First Set of Stats
Attached Image


Second set of Stats (I don't have the delayed Cbet stats though so will need to download them if we need). Positions in the same order.
Attached Image


Lastly nothing on stealing so in here so thought I would do a bit on that. Win rates in there as well as I missed that out earlier.
Attached Image


Posted 8 years ago
Hi w34z3ls great thread and I think it would be EV- not to actually post here.I have been struggling at NL10 and I hope this will give me to finally eliminate costly leaks.

Here's my overall graph:
Attached Image


And here's the positional stats:
Attached Image


Attached Image



From what I understand from the graph my red line needs to be improved dramatically and I feel it is going to be proved in stats, however I don't know where to start myself.


I want to once again say that I really appreciate what you're doing here (and ofc huge thanks to Jon for organising it!) and can't wait to see your comments.
Posted 8 years ago*
Hey W34z3l!

This is 25 and 50NL (15k hands at 50NL)
I've recently implemented a much wider and laggyer BB V SB strategy so I've posted the BB stats separately, filtered to the last 20k hands.
I also very recently increased my BTN open strategy to more like 65% but only for the last 2k hands or so.

My Own Observations

I'm really unsure what my float stat table is like, I haven't really learnt this area yet, could you perhaps expand on this area if necessary?
I need to increase my turn cbet so presumably I need to cbet flop less or bluff more?
I need to fold less to 3bets but I'm really unsure how to expand this range, I currently tend to call AQ+ and pairs.
I definitely need to c/r way more from the blinds, I recently watched your video which should help a tone!
I'm looking to increase my 3bets from the BTN and BB to 8% and 12% respectively.

Overview
Attached Image


Attached Image


Cbets
Attached Image

Attached Image


Floats
Attached Image

Attached Image


Graph
Attached Image


Thanks for the help, I really value your feedback!
Posted 8 years ago
@colly191091

Generally speaking most of the stats without position are not super useful, it's hard to detect problems.

To give you an example of this, our 3bet frequency without position looks reasonable, around 6.3%. But as soon as we check by position we identify some problems. Notice how our 3bet frequency is barely different depending on whether we are in BTN or CO. BTN should be significantly more aggressive than CO for hopefully obvious reasons.

As a rough guide,

MP 4%
CO 6%
BTN 8%

Blinds should be 8% in theory, but I wouldn't be surprised if specifically at 888 network we are better off with a very tight 6% or so. The spot where we should definitely 3bet aggerssively from BB is BB vs SB. Might be worth running a quick filter after this to see what you 3bet frequency is vs a SB open. It should be around 15%, even at 888. Basically, 3 bet like crazy IP but you can be tight OOP just because the games are so fishy here.

4bet ratio is super low at 1%, this is to be expected at these games.

We don't appear to have RFI by position, kinda important.
We don't appear to have call PF 2bet, kinda important. (We just have cold-call PF 2bet, this is not that helpful)
We don't appear to have raise vs flop cbet, kinda important.
We don't appear to have fold to cbet stats on any street, kinda important!

Cbet stats look reasonable, possibly even a shade high for flop cbet. Cbet river success seems high enough that triple barrel bluffing is viable. Maybe even somewhat surprising at this network.

WWSF, WTSD, WSD look reasonable.

Fold to 3bet is super high. Then again, there is a chance you have posted the wrong stat. Please check this is specifically fold-to-3bet-after-raise(2bet), otherwise the stat is useless.

Ok, so basically, we are missing most of the important info to create a helpful review here. Maybe you could try adding the important stuff and reposting?
It seems clear that 3betting strategy needs a little bit of work though and needs to be more reflective of being positionally aware.

Also prolonged swings can happen. Even 70k hands at micros. Will probably happen once every 1 to 1.5M hands though. Unlikely overall but definitely possible, and will happen from time to time if you put in enough volume.



Posted 8 years ago
@thetallpaul

good job at following instructions in the first post.

RFI

EP and MP both look a shade high at 16 and 19%. Do you have a starting hands chart? Does it advise these frequencies? Sometimes we open wider for fish, but usually won't quite be this high. Might be worth checking for one or two bad opens.

CO is high, but probably ok.

BTN - Here is the strange thing, you are overly aggressive with RFI in every position but are suddenly nitting the button up like crazy. We'd expect to see slightly over 50% RFI here. Is 43% what your hand chart advises? Even if it is, we must be missing a ton of steal attempts. Do you have fold-to-steal stats on your HUD?

SB - at around 50%, this is on the aggressive side. Should be 36% by default and we obviously raise super aggro if BB folds more than 60%. Just weird how you are aggro monkey in every position and then a nit on the BTN. Consult your hand charts, find out what is going on.


Fold to 3bet looks fine for these limits.

Call PF 2bet, looks too low at 9.72. We are expecting in the region of 12-13%. The damage seems to be coming in the BB. We have around 13% cold-call in the BB. Should be more like 26%. We are basically just not defending our BB properly. Probably a good idea to check out some vids/articles on blind defense. Other cold-call stats look reasonable, so it's just BB letting you down.

3bet, nice and aggressive, shows reasonable signs of positional awareness. Tbh, unless you are playing high limit games your blind 3bets are through the roof. Again, weird how you are aggro monkey OOP and pretty tight IP. Should be the other way round in vast majority of games. I would consider bringing the blind 3bets down to a healthier 8% and then really focusing on getting that BTN 3bet up to around 8-10%. 6% in CO. 4% in MP. You are probably making the game a lot harder for yourself than it needs to be right now.

WWSF looks low, we'll check that out in a minute with the advanced stats.

Raise flop cbet should be more aggressive, around 15%. This can help your red line.

Fold to cbet stats look reasonable for these games.

Ok so issues with WWSF. First are this float flop and float turn stats. These should be more in the 70% region. Remember that this is an automatic profit spot in the vast majority of games.

Same goes for probe turn and probe river. Stats are too low. We are missing good bluff spots. This is one of the reasons our WTSD is a shade high at 26%. So yeah, basic idea is you need some training on improving bluffing frequency. I think I made some vids here calleding intermediate bluffing spots and advanced bluffing spots. You should probably go an watch these now. Also ramp up 3bet frequency IP and chill out a little OOP, should have some good results.

As we can see, skip cbet and defend by check/raising is pretty much non-existent, so something to think about for the future. You are not thinking much about defending your checks. Not a big issues right now, something for later on.




Posted 8 years ago
Archinator

Just some general pointers regarding your comments. I highly doubt that you want a BB 3bet of 12%. Even if it was not that bad in theory you are going to make the game super tough for yourself while at the same time getting killed by rake. 8-10% BB 3bet should be fine for the games you are playing. Would like to give you advice on not folding so much to 3bets, but I can't actually see the stat anywhere!


RFI - No idea why MP and EP RFI are so similar. They shouldn't be. Your EP RFI is basically through the roof. I don't see any conceivable reason why it would ever reach 19% over 62k hands. Probably good idea to go back to some hand charts for a recap. Other positions look reasonable, although SB is hyper aggressive. I hope you are making good use of fold-BB-to-SB steal stat and not just randomly opening any 2 vs unknowns.

cold-call info looks good and shows a good sense of positional awareness. BB looks slightly underdefended. I doubt this is a huge problem, but we'd ideally expected to see perhaps 26% or so for BB cold-call.

WTSD looks super high. WSD looks too high. WWSF looks low. Sometimes symptomatic of not liking to bluff in good spots. We'll see more in a bit.

We want to increase your 3bet stats in position slightly. BTN can be 8-10%, CO 6%, MP 4%. I really wouldn't worry overly about the blinds at first, but no harm in trying to increase to around 8% or so. Keep in mind that expanding 3bet range IP is easy while expanding 3bet range OOP is difficult. Makes sense to take the easy dollars first.



Cbet T is low as you mentioned. We'd expect to see around 52%. I can't tell you the cause of this through looking at stats only. A common problem is that player have a tendency to check back vulnerable made hands on the turn rather than betting small for protection. Another less common issues is chickening out with good gutshots and flush-draws "just in case we get raised". Something to filter for.

River cbet is super high. Not sure if you bluff a lot or whether this is just symptomatic of someone who is not betting the turn enough. Could be either at this stage. Things will become clear when we fix turn play.

Fold to cbet stats look reasonable. A good fold to R cbet stat is around 60%. Might be worth running a filter to see if you have called 3 barrels with bluff-catchers at any point (usually a losing strategy at your current limits).

Float stats are relevant to your WWSF and red-line. All float bet opportunities are auto-profit spots, so we should be betting more frequently than is theoretically optimal. They should often be around the 70% mark with the exception of river. (We usually get here with a showdown weighted range and protection has no benefit). We can see that you are doing reasonable with flop floats, but turn floats are on the low side. Possibility that we are not betting with showdown hands for protection here when we should be.

Another auto profit spot is delayed cbets. Your delayed turn cbet is 29.5%, and I can't explain how much this makes me want to shoot myself. You don't like free money or something? If you get the opportunity to delayed cbet, always take it unless you specifically have a value-hand and a read. This doesn't include hands "with showdown value", just top pair and better. Bet everything else, always. Delayed turn cbet success is above 60% at these limits.

These leads me to suspect that you might be missing other really important and trivial bluff spots. For example, probe turn and probe river. These were not posted here. But check them. They should both be very high. There is a chance that they are low in your DB and you are missing out on auto-profit.




Posted 8 years ago
Olobury

RFI looks ok-ish by position, but passive on BTN. Probably missing some steal attempts here.

3bet is reasonable but on the passive side. Especially in position. Probably would be a good idea to see how you can spend some time increasing 3bet frequencies to the following numbers.

BTN - 8-10%
CO - 6%
MP - 4%

I think I made a vid here on pokerVIP called advanced 3betting or something. See if you can find it and check it out.

Vs 3bet fold looks fine.

The overall cold-call looks fine, but generally we are not defending our big-blind enough, should at the very least be 23% cold-call, maybe higher.
Might be defending slightly too much by cold-calling in MP. Probably not a big issue though.

Flop cbet stat is around 54.3% which does not take advantage of populatation tendencies. Average opponent is overfolding to cbets and we can exploit this by cbetting aggressively. Even though 54.3% is probably closed to a balanced frequency, it's better to cbet more like 65% in these games. While we don't know that exact reason for the low flop cbet just looking at the stats, a common problem is missing cbets with vulnerable made hands. For example mid and bottom pairs on drawy textures should be cbet IP almost always.

Turn cbet looks low, and this is despite the fact that we should get to the turn with a tighter range than average due to our low flop cbet stat. Basically, cbetting and barreling games needs quite a bit of work and needs to be more aggressive.

Fold vs cbet stats look good.

Regarding your red line comments, why don't you try posting the stats listed under the advanced section of the DB analysis initial post. These are especially for improving red-line.

Your stats in 3bet pots looks reasonable, slightly passive again.

WWSF is very low, and WTSD is high. WSD is a shade high also. These usually indicate a problem with missing bluffs. So maybe try following the step mentioned above and we can hopefully improve your bluffing fundamentals.
Posted 8 years ago
w34z3l: @colly191091

Generally speaking most of the stats without position are not super useful, it's hard to detect problems.

To give you an example of this, our 3bet frequency without position looks reasonable, around 6.3%. But as soon as we check by position we identify some problems. Notice how our 3bet frequency is barely different depending on whether we are in BTN or CO. BTN should be significantly more aggressive than CO for hopefully obvious reasons.

As a rough guide,

MP 4%
CO 6%
BTN 8%

Blinds should be 8% in theory, but I wouldn't be surprised if specifically at 888 network we are better off with a very tight 6% or so. The spot where we should definitely 3bet aggerssively from BB is BB vs SB. Might be worth running a quick filter after this to see what you 3bet frequency is vs a SB open. It should be around 15%, even at 888. Basically, 3 bet like crazy IP but you can be tight OOP just because the games are so fishy here.

4bet ratio is super low at 1%, this is to be expected at these games.

We don't appear to have RFI by position, kinda important.
We don't appear to have call PF 2bet, kinda important. (We just have cold-call PF 2bet, this is not that helpful)
We don't appear to have raise vs flop cbet, kinda important.
We don't appear to have fold to cbet stats on any street, kinda important!

Cbet stats look reasonable, possibly even a shade high for flop cbet. Cbet river success seems high enough that triple barrel bluffing is viable. Maybe even somewhat surprising at this network.

WWSF, WTSD, WSD look reasonable.

Fold to 3bet is super high. Then again, there is a chance you have posted the wrong stat. Please check this is specifically fold-to-3bet-after-raise(2bet), otherwise the stat is useless.

Ok, so basically, we are missing most of the important info to create a helpful review here. Maybe you could try adding the important stuff and reposting?
It seems clear that 3betting strategy needs a little bit of work though and needs to be more reflective of being positionally aware.

Also prolonged swings can happen. Even 70k hands at micros. Will probably happen once every 1 to 1.5M hands though. Unlikely overall but definitely possible, and will happen from time to time if you put in enough volume.





This is great man! Thanks! Really appreciate it!

I'm not too familiar with filtering stats but I'll have a play around when I get home and see what I can get over to you! Smile
Posted 8 years ago
Hello W34z3l,

Good job to do this! Thank you very much.

First some information about my graph. I play NL 5 en at the last 2 months I play NL10.
On the graph the Redline is climbing because i play more from the Button. (Attemp to steal).
Also i play more from the blinds.

At this moment i train for play with 3bets. More agressively!

Greetz. Pokerruudje

Attached Image

Attached Image

Attached Image

Attached Image

Attached Image


Posted 8 years ago
Thanks Weasel, lots to look at for me.

Very interesting actually I'd not spotted my loose opens from EP and MP. CO and BTN are pretty standard for me, SB is a bit loose but I do use steal stats so hopefully that's the only reason. I will tighten up early position opens, or at least investigate why I am deviating from my range. BTN is a bit tighter than my range but only a couple of %, again I'll try and figure out why. I guess it will be when I have an aggro or good player in the blinds but good enough fish to my right so don't want to leave the table. I doubt I have the skill to profitably open 50+% though. Perhaps something to work on when we have our coaching session.

The biggest priority for me though will be BB play and I think that's the focus for the coaching. I 3bet a lot because I don't want to call OOP and then have to x/f all the time when I miss everything. I made the move to that 3bet driven play and had an incredible run with it from about 40k to 75k hands on my chart but I will revisit as its been less effective over the last 100k hands! Again I doubt I have the skill to defend my BB that wide. Currently vs steal I call 14% and 3bet 14%. If I call and bit more and 3bet less and keep the % defend about the same would that make sense? Something like 20% call and 8% 3bet? What do you think?

Again over time I can look at widening them if I can defend well and improve my skill but I rarely semi-bluff enough and I am very inexperienced in X/R. I watched your videos on it and I understand the principle (or at least your explanation of the principle) but in practice I always find a reason to be weak passive. Baby steps though right.

Oh one other thing, what sort of hands do you raise flop cbet with? Currently I use about 50% nuts and 50% non NFD, weak OESD, occasional GS and overs. I call quite a lot of nut draws.

Actions:

Tighten EP and MP
3bet BTN a bit more.
Call BB wider and 3bet less.
Don't be a pussy when they check to me and I have nothing.
Posted 8 years ago
Thanks so much for the feedback Weasel!

Missed Stats
Attached Image

Questions about the Feedback

My SB strategy is wide to take advantage of BB folding tenancies in Rush poker. Or were you referring to post-flop play?
'WTSD is super high', what sort of value should it be and is this possibly because of my lack of turn barrels?
I noticed that calling hands like QT-AT BTN v CO I broke even yet for a 3b I was making about 300bb/100, could it therefore be right to massively increase my 3b to like 12%-15%?
My default UTG RFI is 17.5% for a 2.5x, I then expand if theres weak players at the table with JT--QJo, is that ok or do I need to cut out weak pairs, suited connectors and weak Axs from my default?
Posted 8 years ago
Hi Adam,

Once again, thanks so much for the valuable feedback! I have a list of things to adjust now Laugh

I have configured my report to show the RFI, Call PF 2bet, raise vs. flop cbet & fold to cbet flop + turn as requested in your previous reply

I would appreciate if you could take a look! You the man w34z3l! Smile

Attached Image
Posted 8 years ago
@Pokerruudje,

Ok, overall graph and winrate look great, so something is going right over this sample.

While you state that your red-line is climbing because of your BTN steals, I would say this absolutely not the case. You are actually extremely passive on the BTN, opening just 36% of hands when we could be opening 48% or more. The red line must be the way it is for another reason. Basically there is room to steal a lot more aggressively on boththe BTN and the SB. SB rfi is around 26.3 and should be at least 36%. Open raises from other positions look fine apart from EP where we should be openining significantly tighter. 14% instead of 18%.

3bet game actually looks very good for these limits. There is room to expand your aggression, from the SB for example, but the general idea is that you are positionally aware and this will be one of the contributing factors to your red-line.

Vs 3bet fold is reasonable. Normally I would say it is on the low side, but given you are opening such a tight range in the first place, it is to be expected.


Cold-call numbers look a little strange. Notice how your CC% is basically the same from MP/CO/BTN/SB. It shouldn't be like this. CC should be around 6. So should MP. Maybe CO around 8-9. Btn should be just over 10. Maybe 12. BB cold-call of 17% indicates big blind is not defended enough since the cold-calling frequency should be over 20%.

Cbet stats look ok. Turn and river barrels are actually very high. This will be another contributing factor to your red-line. It probably doesn't need too much adjusting but I wouldn't be surprised to find one or two spots in your DB where you have barreled the turn too aggressively without equity. I could be wrong though, it might just be a function of the tight open-raising range. Naturally if we open wider, our cbetting stats usually get a little lower.

Fold vs cbet stats are ok, apart from fold-to-river cbet which is around 45.9%. This is too low. Should be more like 60%. This is mirrored in your river call efficiency which is an extremely low 1.01. Basically, you are probably making some bad river calls. This will cause your red line to be arbitrarily high for the wrong reasons. In other words, some of the correct adjustments for you right now will likely make your red-line a little lower. W$SD being a shade below 50% also suggests one or too light river calls.

Flop raise cbet is around 10% and could be around 15%. This is another area to focus on maybe a bit later. Bluff-raising flops more aggressively.


Something doesn't feel 100% ok with 3bet pot numbers. Cbetting way too much in theory. Also looks like we are potentially calling down too much vs cbets. I get the general feeling that we are not folding enough in 3bet pots. Certainly makes us tougher to play against, but can be problematic at lower limits since our opponents often enter into 3bet pots with a much stronger range than they should.

Delayed cbet is a relatively neutral 57%, and should be through the roof at these limits. Maybe even above 80%.

Bet vs skipped cbet stats look good, but should be noticeably higher on flop and turn. Grats on getting your river bet vs skipped cbet so high, I'm really wondering if there are some bad plays there with such a high frequency.

Bet turn OOP vs skip is 45.9% and significantly too low. Bet river OOP vs skip is 77.2% and very good.

Ok, great results overall, but a number of issues with the stats.

Posted 8 years ago
thetallpaul:

Oh one other thing, what sort of hands do you raise flop cbet with? Currently I use about 50% nuts and 50% non NFD, weak OESD, occasional GS and overs. I call quite a lot of nut draws.



Probably a good idea is to start raising quite a few more of the nut draws. Anything that can make the nuts by the river is usually a good raise unless the board texture is super dry vs an unknown.

This includes things like backdoor NFDs and nut gutshots. But if we have a draw that's not to the nuts, even if it's a combo draw, usually a good idea to play a smaller pot with it.

Few final things to keep in mind.

1. In single raised pots with 100bb stacks we should be bluffing more often than we value-raise. 2:1 bluff to value ratio roughly speaking.

2. IP strategy can differ noticeably. It's now ok to take things like mid and bottom pairs and use them as bluff-raises, almost like a range merge. We fire safe turns for small sizing and check back river to realise equity vs busted draws.
Posted 8 years ago
@colly191091

RFI probably needs a little tweaking. UTG and MP are a little tight and should be around 13% and 18% respectively. CO is a little too aggressive and should usually be toned down to 28% or so. The numbers are roughly in the right ballpark though.

Raise flop cbet is low. It's around 8% and should be more like 15%. Just means we are not bluff-raising flops as much as we should. Not a big deal at this stage, perhaps something to think about for the future.

No big problems with fold to flop and turn cbet stats. They are both a little bit too high making it profitable for our opponents to cbet very aggressively. But this is not a huge deal at the lower limit games.
Posted 8 years ago
@Archinator

So basically probe turn and river stats are too low. We should be bluffing a lot more aggressively in these spots.

Btw, unless you are the biggest nit I have ever seen, I don't think that's the correct fold to PF 3bet stat. PT4 throws us a bit of a curveball --> we must use "fold to PF 3bet after raise (or 2bet)"

Make sure you don't use the wrong stat on your HUD either, this is a very common problem.

For SB I am talking about preflop. I'm not saying there is necessarily a problem here, so long as your aggression is measured. If you are making good use of the fold BB to SB steal stat then fine, nothing to worry about. Someone could have exactly the same stats and just be opening too wide randomly though. We just have to make sure you are in the first category and not the second.

WTSD should be below 25%. Often a high WTSD can be a symptom of not bluffing enough postflop. Your low WWSF also reflects this idea.

Regarding 3betting, you have the right general idea that increasing 3bet aggression on BTN is probably going to be a lot more useful than increasing 3bets OOP. Sometimes the BB is IP though and we can 3bet a ton from this spot.

For RFI UTG. I get that we can open wider if there are weak players in the blinds. But even if we did this aggressively there is little chance we would ever reach 19%. Pretty good chance you are opening way too wide in some bad spots. Even if we raise wide to exploit fish we are going to hit 15 or 16% absolute max. But 19%.....no way imo.