Flat Betting vs Unit Sizing: Which Strategy Wins Long-Term
Flat betting and unit sizing are the two staple bankroll approaches in sports betting, and people fight about them online constantly. The winner is not who you think. Actually, you might be wrong about both.
Okay so this topic makes people go absolutely nuts online.
Flat betting versus unit sizing. Which one wins? Which one is smarter? Which one is the actually-good pick when you zoom out over a year of action?
Stay with me here, because the answer is kind of a twist.
Claim 1: Flat Betting Is for Beginners
The claim: Flat betting, which is just wagering the same dollar amount on every play, is a starter strategy. Real bettors scale up on their strongest picks.
Reality: Flat betting is actually what most professional handicappers recommend to new bettors and plenty of experienced ones. Why? Because it strips out the biggest leak in sports betting, which is you being way too confident on plays that feel like locks. A flat $50 bet on your 55 percent edge play and a flat $50 bet on your coinflip play will, over hundreds of wagers, reward the edge and punish the coinflip mathematically. Adding subjective confidence tiers usually just lets your worst bias win.
Claim 2: Unit Sizing Is Smarter Because Some Plays Are Better
The claim: Some picks are stronger than others. So you should bet more on the strong ones and less on the weak ones. Unit sizing lets you do that. This is Kelly Criterion stuff.
Reality: This is actually true in theory. In practice, it almost never works for recreational bettors. Kelly sizing requires you to know your true win probability on every bet. Not feel it, not estimate it, not vibe it. Know it. Most bettors massively overestimate their edge on games they love and underestimate variance on games they dislike. Unit sizing without a verified win rate is just feelings-based betting with extra math on top.
Claim 3: Flat Betting Caps Your Upside
The claim: If you are going to beat the market, you need to press hard on your best plays. Flat betting leaves money on the table on the games you actually have figured out.
Reality: This assumes you can reliably identify which games you have figured out ahead of time. The data on closed sports betting account histories, from people like Captain Jack Andrews and the syndicate-adjacent crowd, shows that bettors are terrible at this. Their so-called A-plus plays often perform statistically similar to their B plays. The plays they thought were locks were just the ones they felt strongest about emotionally. Pressing on those is not pressing an edge, it is pressing a mood.
Claim 4: One Percent Unit Sizing Is the Right Answer
The claim: Bet one percent of your bankroll per play, scaling up to two or three on your best. Classic advice from every podcast.
Reality: This is actually pretty reasonable and a lot better than most alternatives. But the one percent number has to be real. A thousand dollar bankroll means ten dollar unit. If you are betting fifty, your bankroll math is cosplay. The real discipline is not in the percentage, it is in honestly sizing the denominator. Most bettors secretly use a bankroll number that is whatever makes their bet size feel normal.
Claim 5: Flat Betting Gives You the Truest Sample
The claim: If you want to know whether you are actually winning at betting or just got lucky, flat betting is the cleanest data set.
Reality: This one is completely correct and wildly underrated. When you flat bet, your ROI number at the end of the year is a real ROI. When you unit bet, your ROI is smeared by all the times you decided a game was a three-unit play and it just happened to lose. Your record looks worse than your skill, or better, and you have no idea which. Flat betting is a diagnostic. It tells you whether you are a winning bettor or a losing bettor. That information is worth more than the theoretical gains from perfect Kelly sizing that you are not actually doing.
Claim 6: The Winner Is Obvious
The claim: Clearly one approach is better and everyone who does the other is wrong.
Reality: The winner over the long run is flat betting for almost everyone, not because unit sizing is bad in theory, but because unit sizing in practice requires you to be honest about your edge, disciplined about your confidence calibration, and consistent about your bankroll math. Most bettors are not any of those things.
If you have tracked two thousand bets and you know your closing line value is positive, then yeah, size up. Until then, flat bet.
That is the actually-wild part. The boring answer is the correct one.