How to study with Rail

On this page

• Follow the loop: learn an idea, drill it, play, then review your own hands.
• Read a grade and the EV loss behind it, and tell a close call from a real mistake.
• Plan short, mixed study sessions instead of long runs of one spot.
• Focus on one format at a time, starting with cash games.

Concepts used: EV, EV loss and how Rail grades, GTO and exploitative play

A loop, not a syllabus

Reading about poker feels like progress, but it only sticks once you use it at the table and look back at what you did. Rail is built around a loop that does exactly that, and each lesson is one turn of it.

• Learn: read a short lesson and answer its checkpoints before the explanation opens.
• Drill: practise the idea on spots Rail generates and grades on the spot.
• Play: take it to your own games.
• Review: upload your hands and let the analysis grade every decision.
• Repeat: your costliest leaks point to the next lesson.

Nothing locks. The course map suggests an order, but you can open any lesson at any time, and your own reports are often the best guide to where to go next.

What a grade measures

Every decision Rail grades is compared with a reference strategy: preflop charts before the flop, Rail's own solver after it. For your exact hand the reference gives each option a frequency (how often it takes it) and an EV (what it is worth on average, in big blinds).

Your EV loss is the gap between the best option's EV and the EV of the one you chose. Rail measures it as a share of the pot, because losing 2 bb matters far more in a small pot than in a huge one. The concept page "EV, EV loss and how Rail grades" goes through it in full.

• Perfect: the reference plays your option within 5 percentage points of its most frequent one, or you lost no more than 0.1% of the pot.
• Good: the reference plays your option at least 3.5% of the time.
• Inaccurate: played less often than that, losing up to 2% of the pot.
• Mistake: losing up to 8% of the pot.
• Blunder: losing more than 8% of the pot.

Study the spots you meet most

Every hand you play starts with a preflop decision, fewer reach the flop, and fewer still reach the river. So a small preflop leak repeats far more often than an exotic river one, and fixing it is worth more.

Start with the spots that come up in every session: opening, defending the blinds, the first bet after the flop. Your reports show which of them cost you the most; the lessons on those come first.

Short, regular and mixed

Twenty focused minutes most days beats one long session a week. Memory holds what you return to after a gap, and short sessions keep the quality of your attention high.

Mix your practice. Ten different spots in a row feel harder than ten copies of the same one, and that is exactly why they teach more: at the table you never know which spot comes next. Rail's drills deal varied spots for this reason.

Stick to one format while you build the basics. Everything here is cash: 6-max and full ring, mostly at 100 big blinds, with deeper stacks and live games later in the course.

Rules of thumb

• Learn, drill, play, review, repeat: one turn of the loop per lesson.
• Judge a decision by its EV loss, not by whether you won the hand.
• Fix the spots you meet every session before the rare ones.
• Short, regular sessions with mixed spots beat long blocks of one spot.

When they break

• A grade compares you with a reference strategy, not with the best play against your actual opponent. A deliberate exploit can grade Inaccurate and still be right.
• Where Rail has no reference for a spot it does not grade it, so a clean report there means "not measured", not "played well".