Bet Tracker & Analytics
Whether the bets you are actually placing are good ones — which is a different question from whether they have won yet, and is answered by a different number.
The screen, annotated
Every column, and what its number actually means.
Closing line value: the tile that answers the question the rest of the screen cannot. CLV on one bet is (the decimal price you got / the decimal price it closed at - 1). So +5% means you took a price 5% better than the market's last word. The tile is the plain average of that across your settled bets THAT HAVE A CLOSING PRICE — not across all of them. It reads \"—\" when none do.
The line of text under the tiles on Analytics, shown only once you have an Avg CLV at all. It is the thesis of both screens: over any sample you will actually reach, profit is mostly variance and CLV is mostly signal. A green Net Profit over 40 bets is a coin that came up heads; a positive Avg CLV over 40 bets is evidence.
Settled bets only, summed. Open bets contribute nothing to it — they are counted under Cash In Play instead — so this number moves when you click a settle button, not when a game ends. On Analytics the same figure heads the tiles and drives the bars below.
Net Profit divided by CASH risked, not by nominal stakes. The two differ the moment a promotional wager is recorded, because a bonus bet risks none of your money and so belongs in neither the numerator as a loss nor the denominator as capital. Same underlying figure on both screens; the tracker rounds it to one decimal, Analytics to two.
Wins divided by wins plus losses. Pushes and voids are excluded from both halves, so this has a smaller denominator than Net Profit's and a different one again from Avg CLV's. Three tiles side by side, three different populations — which is why they can all move in directions that look contradictory.
Two ledger columns that are the same number for a cash bet and only then. Stake is what you put on the slip; At Risk is what actually left your bankroll, which is $0.00 for anything funded as a bonus bet or site credit. At Risk is the denominator of every ROI on the two screens, so it is the column to check when an ROI looks wrong.
Shown as American odds where the bet has them, falling back to the decimal price. Note the asymmetry: the ledger DISPLAYS -105, while the Add bet manually form and the row editor both ask for DECIMAL odds. -105 is 1.9524 there, not 1.05 and not -105.
Per row, green or red, and "—" while the bet is open — unknown, not zero. For a cash win it is the return less the stake. For a bonus bet it is the whole return, because a bonus bet pays winnings only and the token was never yours to lose.
Appended in green under the selection when the bet was logged from an opportunity that carried an edge estimate. It is what the engine thought the bet was worth AT THE MOMENT you placed it, frozen. It is not a prediction about this bet and it never updates.
An orange line under the selection on rows where nobody captured which side was bet. Such a bet cannot be matched to a closing price, so it can never earn a CLV and is silently absent from the Avg CLV average. Opening the hedge finder from the row and picking the right market writes the side back. If the row has no "hedge →" button at all, it carries no event either and cannot be repaired this way.
Three views of the same money. Bankroll is total capital: cash across your book accounts PLUS the stakes riding on open bets. Cash In Play is just the riding part. Available is just the cash. Balances are ones you maintain — TrueEdge never holds book credentials — so a stale figure here means an account needs updating.
Analytics only: the largest peak-to-trough fall in your running bankroll balance, shown negative. It is computed from bankroll entries — deposits, withdrawals, adjustments and settlements — so a bankroll you never seeded reads a smaller drawdown than you lived through.
The three bar panels at the foot of Analytics. Each bar runs left (loss) or right (profit) from a centre line, labelled with profit and that segment's ROI, and the count in brackets after the name is the bet count. Read the count first: a +38.2% segment over four bets is four bets, not a finding.
The three controls that matter
Ignore everything else on a first read.
Funding (the dropdown in Add bet manually, and in the row editor)
Cash, Bonus bet or Site credit. This is the one control that changes what the numbers MEAN, not which rows you see. It sets At Risk to the stake or to zero. At Risk is the denominator of every ROI, and what Promo Profit counts. The hint under the form spells out the consequence as you pick it.
won · lost · push · void · cash out (the buttons on an open row)
Grading is manual, and nothing exists until you do it — Analytics shows "No settled bets yet" no matter how full the ledger is if every row is open. "cash out" asks for the dollar amount the book actually paid, because it is the one settlement whose value is not derivable from the price you took. Click carefully: these buttons disappear once the row is settled.
All / Open / Settled (the chip row above the ledger)
The tiles always describe your whole history; the table below them describes whatever these chips left. Settled is the honest view when you are comparing rows to the tiles. The date chips beside them key off the settled date where there is one, so a bet placed in August and graded in September lands in September. Everything else on that bar — the book chips, and the calendar and Capital by book panels — can be left alone on a first read.
A worked example
From a real row to a placed bet, including the stake.
Three tiles, three denominators — 41, 33 and 39 — and the one that matters is the 33. Taking -105 on lines that close -118 is buying something real. But 41 bets is nowhere near enough to see it in money: at this stake, a normal run of 41 coin-flips swings several hundred dollars — bigger than the whole edge. So -$212.40 is not evidence of anything and +1.82% is. The reverse case is the one to fear: green P&L with negative Avg CLV means you have been paid for bets that were worse than the market, and that stops without warning. Change the method when Avg CLV goes negative, not when the month does.
When it goes wrong
The price moved, a leg was voided, the row vanished. What to actually do.
Avg CLV reads "—", or it covers far fewer bets than you have placed.
This is the expected failure, not a bug, and it has three causes you can tell apart. A bet added manually carries no event. The form sends book, selection, odds, stake and funding — nothing that identifies a fixture. So no closing price can match it, and it will never have a CLV. The row tells you which of these you are looking at, in orange under the bet. \\\"No game linked\\\" means nothing will ever grade it, and no closing price can ever match it. \\\"Side not recorded\\\" means the fixture is there but the side is not. Open \\\"hedge →\\\" on that one, pick the market once, and the side is written back. And a bet on a market TrueEdge captured no close for is simply not measurable; nothing you do to the row will fix that one. The rule: log from an opportunity row wherever you can. Use the manual form only for bets the board does not reach — and know those rows are invisible to the number you are reading. Before you trust an Avg CLV, count how many rows are behind it.
You clicked the wrong settle button. A won bet is now graded lost, or you cashed out and typed the wrong dollar amount.
Click \"re-grade\" on the settled row. It opens the same panel as \"edit\" does on an open bet, with a row of results at the bottom and a \"Settled on\" date beside them. Picking the right result re-settles the bet. The server REVERSES the prior settlement in your bankroll ledger and books the new one. So the balance comes out right, and both the mistake and the fix stay in the history. Fix the date at the same time if the grade was late. A correction made on Sunday to a Friday bet would otherwise file Friday's money on Sunday. A wrong cash-out amount is the same act: re-grade as cashed out and it asks for the amount again. What you still cannot change is the money the bet was made of: the stake, the odds, the funding type and the book. That money has already moved through the ledger, so those four are fixed and the panel says so. Everything else still saves, including which side it was on and which game it belongs to. Linking a settled bet to its game is worth doing. It is what lets the bet get a closing price and show up in the by-sport breakdown.
Analytics shows "No settled bets yet" although the tracker has rows in it, or the tracker itself is showing bets you never placed.
Two different things. Analytics counts SETTLED bets only, so a ledger full of open rows produces that empty state correctly — grade something and it fills in. Bets you do not recognise mean you are looking at the demo ledger. Without the bet tracker feature, the tracker shows an example with an orange "Example data" banner — an empty screen cannot show you what the tool does. None of it is yours, none of it is saved, and your own ledger is untouched underneath it. Analytics is not available at all in that state, which is the quickest way to tell which of the two you are in.
Common mistakes
The specific ways people lose money with this tool.
Judging the method by Net Profit. Forty bets in, the number is green, so the approach works — or it is red, so it does not, and it is time to try something else.
You will quit a method that was working and adopt one that was not, and you will do it with total confidence both times. At $200 a bet, 40 results swing several hundred dollars on luck alone. That is more than a real 2% edge pays over the same 40 bets. So the sign of the P&L is close to a coin flip however good the bets were. Every hour of work that went into finding the edge is discarded on the strength of a number that was never measuring it.
Use Avg CLV as the scoreboard and Net Profit as the consequence. Decide in advance what would make you stop. "Avg CLV below zero across 100 settled bets with a closing price" is a rule you can apply. Until it trips, keep placing and keep sizing the same. Check the tile monthly rather than daily; it is an average and it moves for the same reason P&L does when the sample is small.
Reading Avg CLV as a fact about your betting, without checking how many bets are behind it.
A flattering number computed over the wrong population. The average covers only settled bets matched to a closing price. So if the bets you log by hand are your hunches, and the ones from the board are the engine's picks, the tile is scoring the engine and calling it you. A +3% Avg CLV over 9 of your 60 bets is not evidence about the other 51, and nothing on the screen says so.
Treat coverage as part of the reading. Filter the ledger to Settled and count the rows, then compare with what is plausibly behind the tile. Rows with the orange "side not recorded" line, and rows with no "hedge" button, are the ones definitely excluded. Push the coverage up by logging from opportunity rows rather than typing bets in after the fact, and repair the legacy rows once each through the hedge finder.
Leaving Funding on Cash when logging a bonus bet or site credit, because the stake box says $100 either way.
Every figure on both screens goes wrong in the same direction. A losing $100 bonus bet recorded as cash books a $100 loss you never took, so Net Profit is $100 light per losing token. That $100 also joins Cash Risked, inflating the denominator of ROI on Risk, and Promo Profit sits at $0.00 reading "0 promotional bets settled" while you are running a promotion strategy. The error compounds quietly across a promo-heavy month into a tracker that says you are losing money you are in fact making.
Set Funding as you type the bet, and read the hint under it. For a bonus bet it says no bankroll is debited and a win pays winnings only. That is exactly the arithmetic. If you catch it later, the row editor behind \"fix\" changes funding on any open bet. Once it is graded that door is shut, so the check is worth making before you settle: At Risk should read $0.00 on every promotional row in the ledger.