What Time Do You Place Your Bets - And Is That the Worst Hour for Your Decision-Making?

[SharpEddie47]

Added a timestamp column to the spreadsheet three seasons ago. Not something I'd thought to track before. Assumed the hour was irrelevant to the analytical quality of the decision.

The data disagreed.

My bets placed between 6am and 2pm: CLV positive at the expected rate. P&L consistent with the edge the model produces.

My bets placed between 9pm and midnight: CLV slightly positive but noticeably lower. P&L positive but below the daytime rate.

My bets placed after midnight: I have eleven of these across three seasons. I didn't know that number until I added the timestamp column. The CLV on those eleven: negative. Every single one.

The decisions felt defensible at the time. I had reasoning for each. The reasoning was apparently worse than it felt.

The pattern I didn't notice from the inside: I was doing my primary analysis in the mornings, identifying qualifying selections, and then sometimes revisiting the markets in the evenings. The evening revisits occasionally produced additional bets that weren't in the morning's output. Those bets were performing below the morning selections.

The late-night eleven were the worst version of this. Analysis done during a period when analysis is genuinely worse, feeling like the same quality as morning analysis.

The question: does anyone else have timestamp data. And for those who don't, what hour do you think your bets concentrate in and do you think that's the right hour.

 

[Klaus]

Have tracked timestamp data since building the current model.

The pattern is structured by design rather than discovered by analysis.

Evening window: 20:30 to 22:00 German time. Children are in bed. The house is quiet. This is the only period I place bets.

This wasn't a decision based on circadian research. It was a practical constraint that turned out to have beneficial side effects.

No bets placed under time pressure during the working day. No bets placed after 22:00 when fatigue is accumulating. No bets placed before the evening review of the day's model outputs.

The structured window: functions as a cooling mechanism between analysis and execution. The bet that seemed urgent at 14:00 has four hours to be reconsidered before the window opens.

The bets that survive the four-hour gap tend to be the better ones.

The bets that felt urgent at 14:00 and still feel urgent at 20:30: those I've noticed tend to be the ones with genuine model support.

The bets that felt urgent at 14:00 and no longer feel urgent at 20:30: I've stopped asking why they passed.

 

[FadeThePublic]

The market timing angle is worth separating from the personal decision-quality angle because they're different problems.

Opening lines: typically released the morning before the game for American sports. These lines are set by the book's model before sharp action has shaped them. Early movers have the most opportunity to find mispriced numbers.

Line settlement: by game time, sharp action has driven prices toward efficiency. The retail bettor placing bets an hour before kickoff is getting a more efficient price than the bettor who moved at line release.

The optimal time to bet from a pure market-efficiency standpoint: early, when the line is freshest. Which coincidentally is morning, at least for daytime games.

The worst time from a market perspective: immediately before the event, when the line has absorbed all available sharpening.

The problem: most casual bettors place bets in the evening because that's when they have time to think about it, and often right before or during the game because that's when the game feels real.

Those are the hours when the market is most efficient and when personal decision-making is most fatigued.

The double disadvantage of the 9pm bet: worse prices and worse decisions simultaneously.

 

[TaffyTipster]

Saturday afternoon is my peak betting hour.

Which is also, frequently, my second or third pint hour.

The correlation between betting and drinking isn't planned.

It just happens that Wales are usually on in the afternoon and the afternoon is when people are in the pub and the pub is when people have a drink or three.

The bets I place at 3pm Saturday in the pub while the match is on.

The bets I place at 11pm Saturday when I'm home and there's still a late match available.

Those are different activities dressed as the same activity.

The 3pm bets: I care about the sport. There's a specific reason for the bet beyond needing something to happen.

The 11pm bets: the Saturday has wound down. The results are in. But the markets are still open and something about that feels like an invitation.

Bronwyn goes to bed at half ten usually.

The correlation between Bronwyn going to bed and the quality of my late betting is something I've noticed without tracking.

 

[ParlayPrincess_88]

Sunday afternoons for me. NFL window, Chiefs game, the whole day has a shape around it.

Those bets are the ones I feel best about. I've been thinking about the games since Thursday, I know the matchups, I have actual reasons for the parlay legs.

The bets I'm least proud of are the ones placed Sunday night after the early games have finished and the late game doesn't involve the Chiefs.

Something about having watched seven hours of football makes me feel like I understand things I don't understand.

The expert feeling that comes from watching a lot of something isn't the same as actually having an edge on it.

But at 10pm Sunday it feels the same from the inside.

 

[DublinDegen]

the late-night hours were the most destructive ones for me...

not because the analysis was worse, there was no analysis...

but because the specific quality of the late night changed what was available in terms of reasons not to bet...

during the day there are interruptions... other things to do... the phone rings... a colleague messages... you have somewhere to be...

at 1am in a dublin flat: none of those things exist...

just the open market and the specific silence that needed filling...

the 1am bet wasn't placed because something looked good...

it was placed because nothing else was happening and the market was still there and i was still awake and those three things together were apparently sufficient...

the hour didn't change the quality of the decision...

it changed the availability of reasons to not make it...

which is a different problem from what eddie's describing but it produces the same entry in the record...

just a bet placed after midnight...

 

[SharpEddie47]

Conor separating the analytical quality problem from the availability-of-reasons-to-stop problem.

These are genuinely different mechanisms producing the same outcome: a late-night bet that underperforms.

For me: the analysis is present but degraded. The reasoning feels sound and isn't quite.

For Conor's version: the analysis isn't present at all. The bet fills a vacancy rather than expressing a probability view.

Both produce a timestamp after midnight.

The timestamp data doesn't distinguish between the two causes.

Which is one reason the timestamp data is useful to know but insufficient to act on alone.

 

[ThePuntingProf]

The question of time of day and decision quality has an interesting relationship with what chronobiologists call the circadian rhythm of executive function, which is the cluster of cognitive abilities that includes judgment, impulse control, working memory, and probabilistic reasoning, and the research is fairly consistent in showing that executive function peaks for most people in the late morning, typically between nine and eleven in the morning, declines moderately through the early afternoon, and then either recovers partially in the early evening or continues declining depending on individual chronotype and sleep debt, the implication for betting which I find myself thinking about more carefully now than I did earlier in my career is that the period when most recreational bettors place their bets, which is the evening, coincides with a period of declining executive function for most people, while the period of highest analytical quality, the late morning, is typically when people have other things to do and aren't thinking about betting at all, the exception being people like Eddie who have structured their betting activity to occur closer to the analytical quality peak, and what strikes me about this is that the structural disadvantage of the recreational bettor relative to the professional isn't only about analytical frameworks and sample sizes, it's also about when the work gets done, the professional trader or the serious systematic bettor does their primary analysis during their peak cognitive hours and places bets within a structured window, the recreational bettor does their thinking during declining cognitive hours while simultaneously managing the emotional experience of watching sport, Margaret had a specific habit of never making any significant decision in the evening if she could make it in the morning, she said the morning version of any decision was almost always better than the evening version, she applied this to everything from financial decisions to whether to have a difficult conversation, I didn't apply it to betting for years after she told me and I'm not entirely certain I apply it consistently now.

 

[TaffyTipster]

Prof's Margaret being right about morning decisions versus evening decisions.

That's just wisdom that applies to everything.

The bets I've regretted most: placed after 10pm.

The bets I've felt most confident about the following morning: placed before 5pm.

The bets I placed between 10pm and midnight: usually based on something that felt certain at the time and looks different at 8am.

Haven't tracked it.

But the feeling of reading my overnight betting history in the morning.

That feeling has information in it.

 

[oli_sussex]

Most bets are placed during the worst hours for both decision quality and market efficiency simultaneously.

This isn't coincidental. The time people have available and the time markets are most liquid are the same hours.

The market's efficiency peaks when attention is highest.

Attention and cognitive quality aren't the same thing.

 

[CoachTony_Bets]

The coaching schedule creates a specific natural experiment on this.

Game days: I'm completely focused on the team from 6am through the final whistle, usually around 9 or 10pm. Any betting decisions on other games during a game day I've tracked specifically.

Those bets are worse than my normal average. Not by a lot. By enough to notice.

The mental resources that should be going toward analytical quality are going toward coaching. What's left for betting analysis is the remainder.

The bets placed at halftime of my own game while I'm supposed to be in the locker room addressing my players: I've stopped doing this. It took longer than it should have to stop doing it.

My son plays on my team now. He's fourteen. The halftime bets stopped when I realised there was a version of the story where he tells it later and it's not a story about his father's betting discipline.

 

[DublinDegen]

tony's halftime story...

the thing about the specific hour being a symptom rather than the problem...

1am wasn't the problem for me... it was where a different problem expressed itself...

the absence of anything else to do at 1am was a problem that predated the betting and would have needed addressing whether betting existed or not...

the bet was the available response to a problem it couldn't solve...

but you could only see that clearly when the 1am hours started feeling different...

which they do now, most of the time...

the markets are still open at 1am...

but the reasons to be awake at 1am have changed...

 

[RajaSingh475]

Added a timestamp column to the spreadsheet three seasons ago. Not something I'd thought to track before. Assumed the hour was irrelevant to the analytical quality of the decision.

The data disagreed.

My bets placed between 6am and 2pm: CLV positive at the expected rate. P&L consistent with the edge the model produces.

My bets placed between 9pm and midnight: CLV slightly positive but noticeably lower. P&L positive but below the daytime rate.

My bets placed after midnight: I have eleven of these across three seasons. I didn't know that number until I added the timestamp column. The CLV on those eleven: negative. Every single one.

The decisions felt defensible at the time. I had reasoning for each. The reasoning was apparently worse than it felt.

The pattern I didn't notice from the inside: I was doing my primary analysis in the mornings, identifying qualifying selections, and then sometimes revisiting the markets in the evenings. The evening revisits occasionally produced additional bets that weren't in the morning's output. Those bets were performing below the morning selections.

The late-night eleven were the worst version of this. Analysis done during a period when analysis is genuinely worse, feeling like the same quality as morning analysis.

The question: does anyone else have timestamp data. And for those who don't, what hour do you think your bets concentrate in and do you think that's the right hour.

i think you need to try at afternoon from 2 pm to 4 pm

 

[FadeThePublic]

The specific hour has a market angle that Conor's situation doesn't have but that matters for anyone approaching this analytically.

Asian markets operate on a completely different time cycle. The Pinnacle prices on European football often get sharpened overnight relative to European time by Hong Kong and Singapore sharp action.

The European bettor who wakes up to significantly moved lines: the overnight price sharpening has already happened without them.

The European bettor who stays up until 3am watching Asian line movement: probably not making good decisions at 3am, but they're seeing the price formation in real time.

The time zone disadvantage for European bettors in Asian-influenced markets is real and structural. It doesn't have a clean solution.

But it's a different problem from decision quality. Both matter. They're not the same thing.

 

[Klaus]

The structured evening window I described addresses the decision quality problem.

It doesn't address the time zone and market timing problem Fade raises.

These are separate issues.

My window: 20:30 to 22:00. This is optimal for German domestic football markets where significant Asian sharp action isn't the primary factor.

For markets where Asian money is the primary efficiency mechanism: the German evening window is probably not the right window.

I have stayed deliberately inside Bundesliga markets partly for this reason. The markets I understand best are also the markets where my time zone is least disadvantaged.

The decision to stay within my specialty has time zone benefits I didn't originally calculate.

 

[SharpEddie47]

Klaus identifying an accidental advantage from specialisation.

The bettor who stays in their specific markets has consistent timing relative to those markets.

The bettor who ranges across multiple sports and geographies is dealing with different market timing rhythms for each.

Consistent timing advantage in your home market plus better analytical depth: two compounding benefits from the same decision to stay specialized.

 

[ParlayPrincess_88]

The Sunday afternoon pattern I described as my best hour.

I've never asked myself why it's the best hour versus why other hours are worse.

Reading this thread: the Sunday afternoon has structure around it. I've slept. I've had coffee. The Chiefs game has a specific start time I've been building toward.

The Sunday night bets after the early games: I'm tired. I've been watching football for six hours. The markets are still open.

The markets being still open is not a reason to bet.

I say this knowing I'll probably ignore it next Sunday.

 

[TaffyTipster]

The markets being still open is not a reason to bet.

That's the sentence.

For me it's Saturday night.

For Conor it was 1am any night.

For Eddie apparently it's midnight during NFL season.

The market's continued existence as a sufficient reason to place something.

The discipline that prevents this isn't about analysis quality.

It's about whether you've decided the market being open is an event that creates an obligation.

It doesn't.

But it feels like it does.

 

[DublinDegen]

the market's continued existence as a sufficient reason...

that's the precise description of what the late-night hours were...

not "there's something good available"...

just "there's something available"...

the distinction between those two things: the entire gap between betting as analysis and betting as compulsion...

and from the inside at 1am that distinction was completely invisible...

 

[CoachTony_Bets]

The 1am window for Conor and the midnight eleven for Eddie and the Saturday night post-Bronwyn bets for Taffy.

Different hours. Different causes. Different levels of seriousness.

The common structure: a time period when the thing that would normally create resistance to placing a bad bet is absent.

During the day: obligation, schedule, other people, alternative activities.

Late at night: none of those things.

The late-night bet isn't only a decision quality problem. It's a what-else-is-happening problem.

The absence of alternatives is the vulnerability. The bet fills the absence.

Improving decision quality at midnight doesn't help if the real issue is that midnight has no structure.