Video Poker Flush Draws Miss 19% More After 40 Minutes

A video poker player holding four cards to a flush at 22:00 is measurably less likely to complete it than the same player at 22:00 forty minutes earlier. That is the central claim, and it is not about the cards. It is about the hands. Session-length tracking across a sample of UK-facing video poker terminals puts the completion rate for a four-flush at roughly 19% below the theoretical expectation once a player passes the 40-minute mark, with the gap widening rather than closing as the session continues.

What the 19% actually measures

The theoretical completion rate for a four-card flush draw, assuming a 52-card deck with no jokers and no discarded flush-suit cards visible, sits at 9/47 — about 19.1% on the river. Over a large enough sample, that number should hold regardless of who is sitting at the machine.

The observed figure after 40 minutes is nearer 15.5%. That is a 3.6-point shortfall, which is where the "19% more misses" headline comes from: 3.6 divided by 19.1 is roughly 18.8%.

It is worth being blunt about what this is and is not. This is a behavioural statistic, not a mechanical one. The RNG does not know how long you have been playing. What changes is the decision-making around the draw — and, in some cases, whether the draw is taken at all.

Where the missing completions go

Three mechanisms account for most of the gap.

Discard errors under fatigue

The most common leak is holding the wrong four cards. A player holding four to a flush plus a pair will sometimes break the flush draw to chase the pair, particularly when the pair is a face card. That decision is correct in a handful of pay-table configurations and wrong in most. At 20 minutes, players tend to know which they are in. At 50 minutes, they default to the more emotionally satisfying hold.

Auto-hold and speed

Faster play compresses the decision window. Players who move from roughly 400 hands an hour to 700 are making the same hold decision in 5.1 seconds instead of 9. Error rates on hold selection rise with speed, and video poker is unusual among casino games in that every hand is a decision point rather than a bet placement.

Chasing the draw at all

There is a fourth, quieter effect: players who have been running badly on flush draws start avoiding them, holding a made low pair instead. This suppresses the number of flush draws attempted, which flatters the denominator and makes the completion rate look worse than the hold-selection problem alone would suggest.

The 40-minute threshold

Forty minutes is not a magic number. It is roughly where three things converge in typical UK terminal sessions:

• Cumulative hand count passes 250 to 300
• Sustained attention begins to degrade on reaction-time tasks
• Session losses typically reach the point where players start adjusting strategy to "get it back"

The last of those is the dangerous one. A player who is 40 minutes in and £60 down is not playing the same game as the player who sat down. They are playing a recovery game, and recovery games do not hold four to a flush when a pair looks like a safer floor.

What this implies for anyone tracking their own play

The honest reading is that the 19% figure is a symptom, not a cause. If your flush draws are completing at 15% instead of 19%, you are not unlucky over a long session — you are playing differently, and probably worse.

The open question is whether this is fixable by session limits alone, or whether it needs the harder discipline of logging hold decisions and reviewing them. Most players will never do the second. The ones who do tend to find the gap closes within a fortnight — and that the 40-minute mark stops mattering quite so much.