How to Track Sports Bets: The Log Columns That Matter

To track sports bets well you need seven columns: the date you placed it, the event, the selection, the odds you took, your stake, the result, and the closing line. The first six tell you what happened. The seventh tells you whether it should have happened, and it answers that question roughly 150 times faster than your profit column can.

That gap is the whole reason this guide exists. Here is what to log, how to read it, and where a spreadsheet stops being enough.

The Interactive Bet Log

Below is a log of 8 settled wagers across 9 legs. Switch a column off and watch which readouts go dark, and which one quietly goes wrong instead.

The bets in it are made up for this guide. They are not anyone's real results, not a typical stretch of betting, and not a forecast of what any log will show. They exist to demonstrate what each column does.

Every number in it is computed by the same functions the SmartStake bet tracker grades and scores with, so it behaves the way a real log does. Two things in it are worth carrying into the rest of this guide.

The two headline numbers disagree. The record is 2-5-1 and the log is down $44.76, an ROI of −11.19% on the $400 actually at risk. Yet 6 of the 9 legs were taken at a better price than where the market closed, for an average closing line value of +2.42%. One of those two readings is signal and the other is mostly noise, and it is not the one in dollars.

Switching off the free bet flag does not blank a number, it changes one. ROI drops by 5.22 points, because the log starts charging you a $25 stake you never actually risked. A missing column that leaves a gap is an inconvenience. A missing column that silently returns a plausible wrong answer is the expensive kind.

What to Log on Every Bet

These are the columns that earn their place. The first three are bookkeeping, the middle four are arithmetic, and the last three are what let you slice the log later.

Two of these get skipped almost universally, and they are the two that matter most later.

The timestamp. Not the game date, the moment you placed the bet. Lead time is one of the more revealing splits in a log, because an edge that only exists 10 minutes before kickoff is a different business from one that exists on Monday.

The closing line. More on this below, because it is also the one column you cannot realistically fill in yourself.

Reading Profit at a Small Sample

Look again at the log above. Down $44.76, an ROI of −11.19%, a losing record. Every instinct says the strategy is broken.

Here is the problem with that instinct. Take a bet at −110 and assume, for the sake of the arithmetic, an edge of 2%. That is a figure chosen to work through, not a typical or expected result, and nothing here says you will have an edge at all.

On that assumption the bet wins about 53.4% of the time, and its return is either +90.9% of your stake or −100% of it. That works out to a swing of about 95% of stake per bet, against an edge of 2%. The noise is roughly 48 times the size of the thing you are trying to measure.

Run that through the same one-sided significance test the SmartStake performance dashboard uses on its edge chart, and reaching 95% confidence that a 2% edge is real takes about 6,136 settled bets. At 20 bets a week, that is closer to six years than to a season.

So over 8 wagers, or 80, or 800, your profit column is not telling you whether you are betting well. It is telling you what happened, which is a different and much less useful thing. Here is where the closing line earns its column.

Closing Line Value as the Deciding Column

Closing line value, or CLV, compares the odds you took against the market's final price before kickoff, with the bookmaker's margin stripped out. It grades the decision rather than the outcome:

Your decimal price multiplied by the fair closing probability, minus one. That is the expected value of your price measured at the market's best estimate of the truth. The full explanation and a calculator live in their own guide, along with why removing the vig from the close is not optional.

What matters for your log is how much quieter that number is. A win or a loss swings 95 points. CLV moves on the order of single digit points from bet to bet, because it measures a price against a price rather than a result against a stake.

Feed the same significance test an average CLV of +2.4% and a 9 point bet to bet swing, both assumed rather than measured, and it clears the same 95% bar at about 39 logged legs. Drag the sliders below to see how the count moves with your own numbers.

39 legs against 6,136 bets. Same test, same confidence bar, roughly 157 times less data. That is the entire argument for the closing line column, and it is why a log without it can only ever tell you about the past.

Two honest limits, though. CLV is a leading indicator, not a promise: a bet can beat the close and still lose, and a bettor with good CLV can still finish a month down. And it only works if you compare against a de-vigged close rather than a raw book price, which is a real piece of work covered in the de-vig playbook and handled by the de-vig calculator.

Where a Bet Tracking Spreadsheet Stops Scaling

A bet tracking spreadsheet is a genuinely fine place to start, and it will carry a few hundred straight bets without complaint. It stops scaling on five specific things, and none of them are about row count.

Dollars are per wager, rates are per leg. The log above holds 8 wagers but 9 legs, because one of them is a two-leg parlay. Profit, stake and ROI belong to the slip. CLV belongs to each leg, since each leg had its own price and its own close. A sheet with one row per slip cannot average CLV per leg, and a sheet with one row per leg double counts your stake.

The ROI denominator is stake at risk, not stake. A free bet costs you nothing, so it does not belong in the bottom of an ROI fraction. That is the 5.22 point gap the widget's free bet toggle opens up, and it moves the wrong way, flattering nothing and understating your actual return on risked money.

A pushed parlay leg does not simply void. It drops out of the slip and the payout recomputes from the surviving legs at their combined odds. Any formula shaped like IF(result="win", stake*odds, 0) gets that case wrong, and it gets it wrong quietly.

Pending bets have to stay out of the profit column. A settled-only view is easy to state and easy to break, and a half-graded parlay sitting in your totals will move them around for reasons that have nothing to do with your betting.

You cannot fill in the closing line by hand. This is the real wall. The closing line has to be captured across books at the moment the market closes, then de-vigged into a fair probability. By the time you sit down the next morning, the number is simply gone. No amount of spreadsheet discipline recovers it, which is why the single most valuable column is the one a spreadsheet structurally cannot have.

Reviewing the Log

A log you never read is just data entry. A useful review has a shape.

Read CLV before profit. Profit over a month is a variance report. Average CLV over the same month is a read on your prices, and it is the one that should change what you do next.

Slice it, do not just total it. The interesting question is never "am I up" but "which parts of this are working". Split by market, by book, by lead time, and by your own tags. An overall +1% CLV hiding, say, a +4% segment and a −3% segment is two strategies wearing one number.

Treat a persistently negative segment as an answer. If a market shows negative CLV across a real sample, the log has told you something specific and actionable: stop betting that market. That is the log doing its job, and it is worth more than the same information arriving through six months of losses.

Watch where the limits arrive. Your book column will show you which accounts started refusing stake, which is worth knowing before it becomes the reason your volume fell off.

None of this makes any individual bet more likely to win. It makes you better at telling a good process apart from a lucky one, which is the only thing a log was ever able to do.

Moving to a Sports Betting Tracker

At some point the closing line column is the thing forcing your hand, because it is the one you cannot maintain yourself. A sports betting tracker earns its place by capturing the close for you and scoring every leg against it automatically.

The SmartStake Bet Tracker does that, and its Performance page is where the log turns into the splits above: a cumulative profit chart, a CLV summary, the edge significance read from earlier in this guide, calibration, and breakdowns by market, odds range and lead time. CLV scoring shipped in beta and runs on a de-vigged consensus close by default, with other modes available if you would rather compare against the best available price or the book you actually bet.

If you would rather weigh the options first, the comparison of betting analytics tools covers what the alternatives do and do not capture.

Frequently Asked Questions

What should you track for sports betting?

Track seven things on every bet: the date and time you placed it, the event and market, the selection and line, the odds you took, your stake, the result, and the closing line. The first six reconstruct what happened. The closing line is the one that tells you whether the price was any good, which is the part that carries forward.

Is a bet tracking spreadsheet good enough?

A spreadsheet handles a few hundred straight bets fine. It breaks on four things: parlays, where dollars belong to the slip but rates belong to each leg; free bets, which risk nothing and so do not belong in an ROI denominator; pushed parlay legs, which drop out and force the payout to recompute from the survivors; and the closing line, which you cannot fill in by hand because the market has to be captured at kickoff and stripped of vig.

How many bets do you need before your results mean anything?

Measured through profit, a great many. Assume a bet at −110 and an edge of 2%, a figure chosen to work through rather than a typical result. Its per-bet return swings about 95% of your stake, roughly 48 times the size of that edge, so a one-sided 95% test needs about 6,136 settled bets. Measured through closing line value, the same test clears at about 39 logged legs, because CLV is far less noisy than a win or a loss.

Why does closing line value belong in a bet log?

Closing line value compares the odds you took against the market's final price, so it grades the decision rather than the outcome. It settles down fast enough to be readable in weeks instead of years, which makes it the only column in a bet log that tells you something useful early. It is a leading indicator, not a promise, and a bet that beats the close can still lose.

Should you log a parlay as one row or one row per leg?

Both, which is exactly where a spreadsheet struggles. Profit, stake and ROI belong to the slip, because that is what you risked and what settled. Closing line value belongs to each leg, because each leg had its own price and its own close. A log that keeps only one of the two views cannot produce both numbers correctly.

Bottom Line

Log the date, the event, the selection, the odds, the stake, the book, the result and the closing line, flag your free bets, and tag your strategies. The first eight columns reconstruct your history. The closing line is the one that tells you something about your next bet rather than your last one, and it gets there about 157 times faster than profit does.

Start the log this week rather than at the start of next season. The bets you do not record are the ones you will most want back.