Tracking Tools and Spreadsheets

A bettor who does not track cannot answer the only question that matters: is any of this working? Most people avoid building the answer because they suspect what it will say.

Every post in this series has ended with some version of the same instruction: log it. This post is the one that explains what that actually means, field by field, formula by formula.

The reason it matters more than it sounds: without records, a bettor operates entirely on memory, and memory is systematically dishonest about gambling. Wins are vivid. Losses blur. The bet you almost placed becomes the bet you would have won. As covered in Post 10, this is not a character flaw, it is how recall works, and the only defense is a written record made before you knew the outcome.

01Why Tracking Is Non-Negotiable

Four things a log gives you that nothing else can.

An honest record of results. Not what you remember, what happened. Nearly every bettor who starts tracking discovers their actual results are worse than their impression of them.

Closing line value. The leading indicator from Post 9. Results take a season to become meaningful. CLV becomes meaningful in weeks, and only if you recorded the closing number.

Segmentation. Which sports, markets, and books actually produce. Most bettors discover one segment carries them and another quietly bleeds.

Tax substantiation. As covered in Post 83, gross winnings and gross losses are the figures your return needs, and reconstructing them in March is miserable.

02The Minimum Viable Log

If you track nothing else, track these eight fields. This is enough to calculate everything essential.

Eight fields, roughly twenty seconds per bet. That is the entire cost.

03The Fields That Separate a Log From a Ledger

Four more that turn a record of what you did into a tool that tells you something.

Your projected number. From Post 76. Recorded before you saw the market price. This is what makes your handicapping falsifiable rather than merely opinionated.

Closing number and closing price. The line where the market settled. The single most valuable field in the ledger and the one most commonly skipped, because it requires going back after the game.

Best number available at placement. Across your portfolio. The gap between this and what you took measures your line shopping discipline, per Post 77.

One sentence of reasoning. Why you liked it. Not a paragraph. A sentence. Six months later this is the only thing that tells you whether a losing stretch was bad process or bad luck.

Optionally, a field for passes: games you evaluated and did not bet, with your number and the market number. Almost nobody does this, and it is how you find out whether your passes were correct.

04Calculated Fields and the Formulas

Everything below can be computed from the fields above.

Profit in units. For a win at American odds, profit is stake times the decimal profit multiplier. For negative odds, that multiplier is 100 divided by the absolute value of the odds. For positive odds, it is the odds divided by 100. A 1-unit win at -110 returns 0.909 units. A loss is −1. A push is 0.

Return on investment. Total profit divided by total staked. This is the headline number, and it is the one to compare against the 2 to 4 percent target used throughout this series.

Win rate. Wins divided by wins plus losses, excluding pushes. Useful but secondary, because a 55 percent win rate at bad prices can lose money.

Break-even rate. For a price at -110, this is 110 divided by 210, or 52.38 percent. Comparing your win rate against your average break-even rate tells you whether you are clearing the vig.

Line differential. The number you took minus the closing number, signed so that positive means you beat the close. On a spread this is straightforward. On a moneyline, convert both to no-vig probability and compare.

Shopping gap. Best available number minus the number you took. Should be zero. Anything else is a leak with a known fix.

05Units Versus Dollars

Track both, and use them for different purposes.

Units normalize across bankroll changes. A bettor who was betting $25 in October and $60 in March cannot compare those periods in dollars, but can in units. Units are how you evaluate handicapping.

Dollars are what you actually have. They are also what your tax return needs, per Post 83, and they are what tells you whether this activity is worth the hours.

The relationship is simple: units answer whether the process works, dollars answer whether it matters.

06Segmentation: Where the Real Insight Lives

Aggregate ROI tells you very little. The same number can hide a strong segment subsidizing a weak one.

Segment by:

• Sport. Nearly every bettor is better at one than another and does not know which.
• Market type. Spreads, totals, moneylines, props, futures. Prop performance in particular is often wildly different from side performance.
• Book. Which accounts actually produce, per Post 78.
• Timing. Early week versus close to game time. This tells you whether your edge is informational or analytical.
• Stake size. Are your two-unit plays actually better than your one-unit plays? If not, your conviction is not calibrated.
• Favorite versus underdog, over versus under. Simple splits that frequently reveal a systematic bias.

The stake-size segment deserves special attention. A bettor whose highest-conviction plays perform no better than their standard ones has learned that their sense of conviction carries no information, which is worth knowing and is invisible without the data.

07Sample Size and What You Can Conclude

The most common analytical error with a fresh spreadsheet is reading meaning into fifty bets.

This asymmetry is the entire argument for tracking closing lines. Results need a year or more to speak clearly. CLV speaks in weeks. A bettor who tracks only wins and losses is flying on instruments that take twelve months to update.

08Visualizing Drawdowns

One chart worth building: cumulative units over time.

It shows two things a table cannot. The overall trajectory, and the shape of the losing stretches along the way.

Seeing that your best season included a fifteen-unit drawdown lasting six weeks is genuinely useful preparation, because it will happen again and it feels catastrophic in the moment. A bettor who has looked at their own drawdown history handles the next one better than one experiencing it for the first time. We take this apart properly in Post 91.

09Spreadsheet, App, or Custom

Spreadsheet. The right answer for nearly everyone. Total control, every formula visible, no subscription, and easily adapted as your needs change. Google Sheets or Excel both work.

Tracking apps. Convenient, often with automatic import from linked accounts, and faster entry on mobile. The tradeoffs: you are limited to the metrics the app calculates, closing line capture varies in quality, and your records live somewhere you do not control.

Custom tooling. Worthwhile only if you are already comfortable with code and want automated data pulls. The same guidance from Post 76 applies: build it because you need automation, not because it feels more serious.

The honest recommendation: start with a spreadsheet. If entry friction is causing you to skip bets, an app that captures 90 percent of your activity beats a perfect spreadsheet that captures 40 percent.

10Making It Actually Happen

The failure mode is not choosing the wrong tool. It is abandoning the habit in week three.

1. Log at placement, not later. Entering the bet as you place it takes seconds. Reconstructing a week takes an hour and will not happen.
2. Use your phone. A sheet you can open on mobile removes the excuse.
3. Batch the closing lines. Fill in closing numbers once a day rather than per bet.
4. Accept imperfection. A log missing a few closing numbers is enormously more useful than no log.
5. Set a weekly appointment. Fifteen minutes to fill gaps and glance at the totals.

11The Monthly Review

Thirty minutes, once a month, in this order.

1. CLV percentage. How often did you beat the close? Above 55 percent is a real edge. Below 50 is not.
2. Shopping gap. How often did you take less than the best available number? Should be near zero.
3. Segment ROI. Which sport, market, and book produced, and which did not.
4. Stake calibration. Did your larger plays outperform your standard ones?
5. Process compliance. Did you follow your own rules on caps, sizing, and the evaluation sequence?
6. One change. Identify a single adjustment for next month. Not five.

Notice that results ranks fourth at best. A month is not a sample. Process compliance and CLV are what a monthly review can legitimately assess.

12What the Data Will Not Tell You

Worth stating honestly, because tracking can create false confidence as easily as it creates insight.

• It will not tell you a losing month was bad luck. It will tell you whether your CLV held, which is the closest available proxy.
• It will not validate a small segment. Twenty prop bets at 65 percent is noise, however satisfying the number looks.
• It will not identify why a segment underperforms. That requires reading your own reasoning notes and thinking.
• It will not stop you from mining it. If you slice the data enough ways, some slice will look profitable. That is the overfitting problem from Post 76 in a new setting.

13Common Tracking Mistakes

• Not recording closing lines. The most valuable field, skipped because it requires a second visit.
• Tracking dollars only. Units are how you compare across bankroll changes.
• Tracking net only. Gross winnings and gross losses are what your tax return needs.
• Logging after the fact. Memory contaminates the reasoning field, which is the point of the reasoning field.
• Reading meaning into small samples. Under 300 bets, results are close to uninformative.
• Never segmenting. Aggregate ROI hides everything interesting.
• Not logging passes. You never learn whether your discipline was correct.
• Building something too elaborate to maintain. The best ledger is the one still being updated in March.

14The Bigger Picture

There is a reason most bettors do not track, and it is not that spreadsheets are difficult.

It is that tracking converts a pleasant uncertainty into a specific answer, and most people already suspect what the answer will be. As long as results are unmeasured, a bettor can believe they are roughly break-even with some bad luck. A ledger removes that possibility permanently.

That is exactly why it is the dividing line. Everything in this series, every model and framework and workflow, is unfalsifiable without records. A bettor with a beautiful process and no data has a hobby with good production values.

The flip side is worth saying too. A bettor who tracks honestly and discovers a real edge has something very few people have: evidence. Not a feeling about a good season, but a number, across a sample, that says the process works. That is worth the twenty seconds per bet by itself.