Notes #1. What people taught me about ICM
Yesterday I asked why jamming beats calling under ICM pressure. The answers were more useful than my question, so here are my notes, for me and for other beginners. Correct me if I got something wrong.
ICM pressure isn't the same weather for the whole table. Each player has their own, and their own bubble factor.
It grows with the number of chips at stake. It's highest when two deep stacks clash, or a medium stack against a big one. Short stacks and the chip leader feel much less of it.
So a small pot is a way to lower the pressure on yourself. And calling doesn't mean "I only win at showdown": you can still take the pot postflop.
Fold equity under ICM isn't "worth more". There is simply more of it when your opponent is under heavy pressure. That's when it pays to steal wider and push him out of pots.
If I'm the one under heavy pressure, jamming can be a mistake: chips I win are worth much less than chips I lose.
Short version from RoofOfTheWorld: "fold equity". A detailed point-by-point breakdown came from SnowBeaver, thanks.
The main flaw in my picture: I thought pressure was the same for everyone at a given stage. It actually depends on whose stack is against whose. When I review hands, I want to start with one question: how much pressure am I under, and how much is he?
Question: how do you judge an opponent's pressure at the table without a calculator — stack vs average, payouts, something else?
Notes #3. How to fit poker into a day when you have a job
I work in my field, and my whole job is calculations. Poker comes after. Right now my poker day looks like this:
about three hours of study: preflop, spot math, reviewing my own hands
about five online tournaments
live series sometimes
And honestly, I don't know if that's the right ratio.
What I've already heard on this forum:
mrtsz plays a morning schedule and doesn't review hands mid-session: he marks tricky hands while playing and looks at them later, separately
in blogs I see people playing 25–40 tournaments a day, or 900 hands an hour in cash, while also juggling a job or school
As a math guy I'd put the problem like this: you have limited time T. How much goes to playing and how much to studying, to improve fastest? Only playing — I repeat the same mistakes over a bigger sample. Only studying — the knowledge never turns into a skill at the table.
My questions to those who've been through this stage:
How many hours of study a day is a really useful minimum in year one, and when is it too much?
What play/study ratio would you suggest to a beginner: 50/50, 70/30, something else?
Study on playing days, or separate days?
I'd appreciate any experience — I'll collect the answers into the next notes post.
Notes #2. Where I got my own puzzle wrong
Two days ago I posted a pot odds puzzle: 1,500 in the pot, a short stack jams 3,000, we're in the big blind, what equity do we need to call. My answer was 40% (3,000 / 7,500).
Stoma22 corrected me, and he's right. I forgot that in the big blind my 800 is already in the pot. It's no longer my money, it's part of the pot.
The right way, with blinds 400/800 and 300 ante:
1,500 in the pot (400 + 800 + 300)
the opponent jams 3,000
I need to add 2,200, not 3,000
final pot: 1,500 + 3,000 + 2,200 = 6,700
equity needed: 2,200 / 6,700 ≈ 33%
That's 7 percentage points. In the BB it moves a whole bunch of hands from fold to call.
What I took from it:
Same formula: amount to call ÷ final pot. But "amount to call" is only what I add now, not what's already in.
A posted blind isn't "mine" anymore, it's in the pot. Easy to forget when doing it in your head.
Same thing in the SB, just a smaller discount.
Thanks to Stoma22 — corrections like this are exactly why I keep this blog.
Question: what other pot odds mistakes do you see most often from beginners?
Lesson breakdown, MTT #1. Ranges: think in distributions, not hands
The academy in my trainer IPPT has 71 lessons: 37 for MTT and 34 for cash, and the app has an exam on them. I'm starting a series: one lesson a day, alternating — an MTT lesson, then a cash lesson. I put them together for my own study, but I'm a first-year player, not a coach. So I'm posting every lesson for review: where it's oversimplified, where it's wrong, what you'd add.
Key points from lesson 1:
Poker is working with probability distributions. The question isn't "what hand does he have", but "what does his range look like and where did it shift after the last action".
A range isn't a list, it's weights: the same hand can be played different ways at different frequencies.
Three shapes: linear (early-position opens, value 3-bets), condensed (call, BB defense, check-call) and polar (3-bet, check-raise, river bet).
By the river every range moves toward polarization: betting makes sense only for value or as a bluff, and everything in the middle becomes a bluff-catcher.
Every decision starts with one question: where is my hand in my range right now — top, middle or bottom?
The example from the lesson I liked most: BTN opens, BB calls. On A72 rainbow the button has the edge — more aces and overpairs. On 987 offsuit it flips: lots of BB hands there — T9, 98, 65s, straight draws. Same players, same preflop, and the texture decides whose flop it is.
Puzzle from the lesson: flop KQ4 two-tone. CO opens, BB calls. Who has the range advantage, and who has the nut advantage? What c-bet size makes sense and why?
I'll post my answer in the comments tomorrow, but I want to hear yours first. And a question: what's the most oversimplified part of the "three range shapes" picture?
The promised answer to the puzzle from MTT lesson #1 (KQ4 two-tone, CO vs BB). Here's how I understand it — correct me:
range advantage is with CO: his open has lots of Kx and Qx with good kickers (AK, AQ, KJ, QJ) plus AA
nut advantage is with CO too: BB 3-bets KK and QQ more often than calls, while CO keeps them in range; BB has K4s, Q4s, 44, but that's few combos
so a frequent small bet, around 1/3 pot, makes sense
the two-tone board gives BB flush draws, and here I'm not sure: is that a reason to bet a bit bigger, or less often?
Anyone who has looked at spots like this in a solver — how close am I?
Lesson breakdown, Cash #1. Ranges at deep stacks
The academy in IPPT has 71 lessons (37 MTT, 34 cash), and the app has an exam on them. Yesterday was MTT #1; today it's the first lesson of the cash track. The range theory is the same, so I'll focus on what's different in cash.
Cash nuances (from the lesson):
Stacks are deep — 100bb, often 200bb+. More streets and decisions ahead, and a mistake in reading a range "compounds over thousands of hands".
No ICM and no short stacks that "simplify" the game in tournaments: everything comes down to the quality of your model.
At depth hand weights matter more. The lesson's example: AKo on the BTN at 200bb vs an aggressive blind sometimes calls a 3-bet to play a deep pot in position instead of bloating it with a 4-bet.
You build a model of the opponent not only from position and actions, but from the HUD and table history — in cash you see the same players for hundreds of hands.
With many streets, you plan the line with the river in mind: by then every range polarizes.
Puzzle from the lesson: villain check-calls the flop, check-calls the turn and leads 3/4 pot on the river. Is that polarized or a middle range? You have top pair, medium kicker — call or fold, and what does it depend on?
Question to cash players: what else is fundamentally different from MTT when working with ranges that I didn't mention?
Question of the day: MTT or cash if you have a job?
Yesterday Pakemaker gave me advice in my blog: pick one discipline — MTT and cash are very different, and a beginner should focus on one. And with a day job he recommends cash: you're not tied to tournament schedules, you decide when to sit down and when to leave.
I play MTT and I'm staying there for now, but it's a strong argument, so I want to collect opinions.
For cash (as I understand it): flexible time; you can leave any moment; lower variance, results show up faster.
For MTT: risk is capped at the buy-in; one deep run can bring dozens of buy-ins.
Against MTT: a tournament can drag into the night; huge variance, results show only after thousands of tournaments.
Questions:
Where would you start today if you can play 2–3 hours after work?
Can you combine both, or is it really a mess in your head?
If you switched from one to the other — what was the hardest part?
The promised answer to the puzzle from cash lesson #1 (check-call flop, check-call turn, 3/4 pot lead on the river). Here's how I understand it — correct me:
the passive call-call line describes a middle range, and a sudden river bet almost always polarizes it: either a completed draw or the nuts, or a bluff
our top pair, medium kicker is a bluff-catcher here
the math: a 3/4 pot bet, after the call the pot is 2.5 original pots, we add 0.75 → we need 30% equity. So the call pays off if the opponent bluffs at least 30% of the time
so the decision depends on the specific player: does he have enough bluffs in this line. People say that at low stakes there are few such bluffs, and folding against this line is more often right
Cash players — do you agree with the last point?
Lesson breakdown, MTT #2. Range structure: where a range breaks
The academy in IPPT has 71 lessons (37 MTT, 34 cash), and the app has an exam on them. Today is MTT #2 — continuing the topic of ranges.
Key points from the lesson:
Any range can be described with four parameters: density (where most of the hands sit), nut density (how many nuts), air density (how much air) and continuity (whether there are gaps).
Linear — stable but inflexible. It breaks on connected boards like 987 or T98, against a raise and against pressure over several streets.
Condensed — resilient but capped. Big sizes work against it (75%+ and overbets), while a small bet just gives it cheap realization.
Polar — strong but needs balance. You don't counterattack it; you bluff-catch and read blockers.
EV is born where structures meet: hit where the other shape loses stability.
A typical mistake from the lesson: raise-bluffing into a polarized bet. You attack exactly the structure with a strong top and run into the nuts.
Puzzle from the lesson: flop T98 two-tone. CO has a linear opening range, BB has a condensed defense. Who is under pressure and why? Which tool from the lesson would you use?
Question: how much do you actually think in these categories at the table — or is it more for reviewing after the session?
Question of the day: how do you review your hands?
On this forum mrtsz advised me not to review hands mid-session: mark tricky hands while playing and look at them later, separately. I'm trying to build a system like that and want to know how others do it.
Questions:
How many hands do you review after a session — all the marked ones or only big pots?
What do you use: a tracker, a replayer, a solver, or just notes on paper?
What gave you the most growth — reviewing your own hands or others' (streams, videos)?
I'd appreciate any experience — I'll collect the answers into the next notes post.
The promised answer to the puzzle from MTT lesson #2 (T98 two-tone, linear CO open vs condensed BB defense). Here's how I understand it — correct me:
CO is under pressure: his linear range is full of AK, AQ, KQ and overpairs, which lose a lot of value on this board
BB has more of the strongest hands: two pairs, sets and straights from suited connectors and T9/98-type hands that defend the blind
the tool from the lesson for BB is board dynamics plus a raise: check-raising here hits exactly where the linear range is weak
for CO it means betting this board less often and not bloating the pot with an overpair
How close am I?
Lesson breakdown, Cash #2. Range structure — and a gap in my app
The academy in IPPT has 71 lessons (37 MTT, 34 cash), and the app has an exam on them. Today is cash lesson #2, and I have to be honest: while preparing it I found that in the current version its text is the same as MTT lesson #2. So there's no cash-specific part in it yet.
The core is the same for both formats: a range is described by density, nut density, air density and continuity; linear ranges break on connected boards and against raises, condensed ones against big sizes, polar ones against disciplined bluff-catching.
What I'd like to add for cash — and here I need cash players:
At 100bb+ there are more streets, so the shift from linear/condensed to polar happens over a longer hand. Does that change which structure you choose on the flop?
Rake: in small pots does it make sense to avoid "thin" lines with a condensed range?
Regulars and HUD stats: how quickly do you read an opponent's range structure in practice?
I'll add the best answers to the lesson in the app, with credit if you want it.
Question of the day: how many tables at once?
In blogs here I see very different approaches: someone plays 25–40 tournaments a day, someone 8 cash tables at 900 hands an hour, someone says they lost their game when they added tables. So:
How many tables do you play at once, and how did you know it was time to add one more?
What drops first when you add tables — decision quality, focus, or the ability to review hands later?
For a first-year player: fewer tables and more thinking, or more volume and review afterwards?
Lesson breakdown, MTT #3. Combinatorics and blockers
The academy in IPPT has 71 lessons (37 MTT, 34 cash), and the app has an exam on them. Today is MTT #3 — how to turn "it feels like he has it" into an actual count.
Key points from the lesson:
Combos: a pair has 6, a suited hand 4, an offsuit hand 12, any unpaired hand 16 in total. Offsuit hands are three times more common than suited ones.
A blocker is a card that removes combos from the opponent's range. If you hold an ace, AA goes from 6 combos to 3.
Bluff with hands that block the opponent's value and calls, not his folds. Example from the lesson: a monotone spade board — a bluff with the A♠ is much stronger than without it.
When bluff-catching, it's the other way round: you want to block his value and not block his bluffs.
On a short stack any ace is a good jam: it blocks AA and AK. A2o is weak as a hand but strong as a blocker.
A side note from the comments: a reader rightly told me that in poker these aren't "rules" but "strong assumptions" — they can still fail against a specific player.
Puzzle from the lesson: on the river the opponent's value range is two sets (3 combos each) and one straight (8 combos). His bluffs are 12 combos of missed draws. He bets pot, so you need about 33% to call. What share of his range is bluffs, and is your bluff-catcher a call or a fold?
Question of the day: which leak took you the longest to fix?
Every first-year player has a list of leaks. I'm curious how it goes for those who are further along:
Which leak took you the longest to fix?
What actually helped — volume, theory, a coach, reviewing hands, or something else?
How did you know it was really fixed and not just a good stretch?
The promised answer to the puzzle from MTT lesson #3. Value: two sets (3 + 3 combos) and a straight (8) = 14 combos. Bluffs: 12 combos. So bluffs are 12 / 26 ≈ 46% of his range. Against a pot-sized bet we need about 33%, so the bluff-catcher is a call — with a margin. One caveat from the lesson itself: if our hand blocks his bluffs, the share drops, and the call gets worse. Correct me if I counted something wrong.
Lesson breakdown, Cash #3. Combinatorics and blockers at deep stacks
The academy in IPPT has 71 lessons (37 MTT, 34 cash), and the app has an exam on them. Today is cash lesson #3. The base is the same as in MTT #3: a pair has 6 combos, a suited hand 4, an offsuit hand 12; a blocker removes combos from the opponent's range; bluff with hands that block his value, not his folds.
Cash nuances (from the lesson):
At deep stacks blockers decide which hands go into 3-bet and 4-bet bluffs. In MTT the lesson's example was a short-stack jam with any ace; in cash it's preflop aggression at 100bb+.
That's why suited wheel aces (A5s–A2s) are the textbook bluffs: they block AA and AK, so you run into a 4-bet or a dominating call less often, and they keep nut potential.
Same on the river: a missed draw holding a blocker to the nuts makes the opponent fold more often.
The lesson calls blockers a systematic filter for choosing aggression hands, not a fine detail.
Puzzle from the lesson: you want a 3-bet bluff against a CO open. What's the better bluff — A5s or 87s? Explain through blockers to AA/AK.
Question to cash players: do you really 3-bet A5s more often than 87s at your stakes, or does it depend on the opponent?
Question of the day: how do you count at the table?
Exact percentages, or shortcuts like the rule of 2 and 4 for draws? I'm drilling pot odds and keep wondering how precise you really need to be in real time.
What do you count at the table and what do you leave for later?
Which shortcuts do you use?
Was there a moment when a precise count changed your decision?
The promised answer to the puzzle from cash lesson #3 (A5s or 87s as a 3-bet bluff vs CO). Here's how I understand it — correct me:
A5s is better: the ace blocks AA and AK — exactly the hands CO 4-bets or calls with while dominating us. AA goes from 6 combos to 3, AK from 16 to 12
if we do get called, A5s has nut potential: the wheel and the nut flush
87s blocks none of the opponent's strongest hands, so we face a 4-bet more often
At the same time 87s plays well postflop, so if I understand correctly it's more often a call than a 3-bet bluff. How true is that at your stakes?
Hi, I'm Ievgenii.
I've loved numbers my whole life. As a kid I was the one counting everything, even counting elephants to fall asleep. I went to a math lyceum, then studied applied system analysis at Kyiv Polytechnic, and later got a master's in civil engineering. I still work in my field, and my whole job is calculations.
The funny part: I first met poker in a combinatorics lecture at university, and it didn't interest me at all. Years later friends invited me to play, and this time it stuck.
This is my first year, and I'll be honest: I play badly. Like most beginners, I only played strong hands, lost chips anyway, tightened up as far as I could, and still lost almost every tournament. At some point I understood that poker is like anything else people do seriously. It needs discipline and it needs study. I want to prove to myself that I can play disciplined, solid, winning poker.
The hand that started it
I got lucky in a small Saturday live tournament: $50 in, about $1,700 out. Instead of going home, I sat down at a cash table far above anything I should have been playing. It went well. I had built my stack to about $9,000 and was literally stacking my chips to leave.
Then I looked down at KK. I raised. Flop: K 8 8. Kings full.
The player next to me, a very wealthy man who was clearly there to enjoy himself and had already reloaded several times, went all in. I thought for a long time and called. He showed A5 and congratulated me.
Turn: ace. River: ace. Aces full. A third player said he had folded AJ on the flop.
I left with about $700.
Here's the math guy part. When the money went in, I was a 99.6% favourite. Of 990 possible turn and river runouts, he wins 4. The call was right. The mistakes were earlier: sitting at a table my bankroll had no business at, and not leaving when I had already decided to leave. The cards didn't beat me. My discipline did.
Later I found out the exact same runout happened at the very top. EPT Barcelona 2014, €50,000 Super High Roller: Sven Reichardt's KK against Olivier Busquet's A2s, board 8 8 K, then ace, ace. Busquet went on to win the tournament. Reichardt did nothing wrong, and neither did my call.
But that's the difference between MTT and cash. In a tournament you can't leave: your risk is the buy-in you already paid, and the chips are just chips. In cash every chip on the table is real money, and you can stand up whenever you want. I had already decided to. That was my edge, and I didn't use it.
That hand is why I started looking for real answers.
What this blog will be
This blog is my notebook:
hands I misplayed: what I thought at the time and what I should have done
leaks I find in my own stats
the math behind spots I keep getting wrong
progress, honestly, including the weeks it goes backwards
I don't think I can get good alone. I want to grow with this community, and to do that I hacked together a little program: a free offline preflop trainer called IPPT. It's basic, but it will only get better if people use it and tell me what's wrong with it. Feedback, wishes, complaints, all of it. The details and the download are in Poker Software.
And if you see me making a mistake in a hand, tell me too. That's the whole point.
To start: what's the one thing you wish someone had told you in your first year?