Yes, for Every Dart It Records; the Edge Cases Are Undocumented
Autodarts averages only the darts it records. Here is the formula it should match, the undocumented edge cases, and a hand-tally check for LENS and X users.
Yes. Autodarts computes an average from every dart that reaches your account, whether the score came from the free LENS phone-camera mode, the £449 four-camera Autodarts X, or your own finger on the virtual board. What it cannot do is score darts it never saw. Sessions on another board, unrecorded practice and any visit the camera misread or you mistyped sit outside the number. Call it an Autodarts-recorded average. For the darts it does record, the conventional formula (points ÷ darts × 3) gives the same figure you would hand-tally, but Winmau’s product pages say only “averages” and publish no rule for busts, short winning visits or corrections, so those edge cases remain undocumented.
Enter points and darts from your own tally of the same legs; the average and misread sensitivity update beside the inputs.
Recorded Average Checker
Points and darts from your own tally of the same completed legs Autodarts is showing.
Formula: points ÷ darts × 3, the conventional three-dart average (default example: 900 points in 45 darts = 60). Winmau's Autodarts pages say "averages" and publish no bust or short-visit rule, so the implied figure assumes the conventional formula. — = not entered.
What Winmau Says the Platform Tracks
Winmau launched the Autodarts ecosystem at IFA Berlin on 3 September 2026. Autodarts X, the dedicated hardware with four HD cameras, opens for pre-order in October 2026 at £449. Winmau says the platform follows performance throw by throw across match history, including averages, checkout performance and scoring patterns, and its Autodarts X hardware page confirms the app tracks averages and match history when paired with the four-camera unit.
Sports Insight reports that the wider ecosystem provides real-time in-game statistics and historical performance tracking. These are advertised capabilities, not an independent audit of the calculation.
Retailer Dartshopper calls the statistic a three-dart average. The official pages use the broader word “averages” and do not publish a formula or inclusion rules. So the platform tracks an average; the documentation does not prove how every mode calculates it.
Three Ways a Dart Reaches Your Account
Autodarts accepts scores through three routes, described on the Autodarts X experience overview:
| Route | How the score enters | Main data risk |
|---|---|---|
| Manual entry | You tap the segment or type the score | Typing, counting or wrong-segment errors |
| LENS | Phone or tablet camera detects darts | A camera read may need checking |
| Autodarts X, Vantage or DIY | Dedicated cameras identify the segment | Detection errors or disputed reads |
Manual scoring and a limited LENS allowance are free; Autodarts X is paid hardware. The four-camera specification belongs to Autodarts X alone. LENS, Vantage and DIY installations use different equipment and inherit none of its claims.
Each route creates score data. None proves every throw was captured correctly. Manual entry removes camera misreads and adds human ones. Automatic scoring removes the entry burden and leaves detected scores that can still need verification.
One account carries one set of statistics across the supported mobile, tablet and web apps, so recorded data follows you between devices. That does not establish that every mode, practice routine or abandoned session feeds one universal figure, or that games played elsewhere are included. Check which player, match, legs or date range the dashboard is showing before reading any number as your lifetime average.
The Formula, and Where the Documentation Goes Quiet
“Real average” means two different things. One is the arithmetic: points scored divided by darts thrown, times three. The other is a fair estimate of your overall playing level. Autodarts can deliver the first if the inputs are correct. Whether it delivers the second depends on the sample, and several full matches with weak legs included say more than one hot session.
The conventional three-dart average is total points ÷ darts thrown × 3. Under it, 900 points from 45 darts gives 60. Conventional counting treats a bust on the first or second dart as a full three-dart visit, while a winning visit finished with the first or second dart counts only the darts actually thrown, per the standard definition of a three-dart average.
Dartshopper describes the Autodarts statistic as a three-dart average in its overview of camera-based scoring. The Winmau pages say only “averages” and do not document how any mode treats busts, missed doubles, short winning visits, corrections or unfinished matches. The conventional formula is the right comparison point, not proof that every Autodarts mode applies every convention identically.
A correctly calculated match average is still a summary of that match. A 60 across several ordinary sessions describes typical form; a 60 from one short burst does not.
One Misread Visit Moves a 45-Dart Average by 2.67 Points
Calculation accuracy and input accuracy are separate problems. A system can apply the right formula to the wrong score.
Say you hit 60 in a visit and Autodarts records 20. Left uncorrected, the recorded total is 40 points short. Across a 45-dart sample that is 40 ÷ 45 × 3, a drop of 2.67 points in the displayed average from one visit. A mistakenly inflated score does the reverse. The source of the error, camera or keypad, makes no difference to the arithmetic.
Autodarts X lets Plus subscribers request an AI-referee review of a disputed read. That is a useful safeguard and an admission that automatic results sometimes need checking. The public material does not say whether a reviewed or manually corrected score rewrites the historical statistics retrospectively.
Accuracy Claims Belong to the Setup They Were Made For
The DIY documentation says accuracy of up to 99% is possible for its documented three-camera configuration. “Up to” is a ceiling, not a guarantee for every board, camera position, dart or lighting condition, and it should not be transferred to LENS, Vantage or Autodarts X.
Dartshopper reports a similar 99% product claim for Autodarts Vantage under normal conditions and notes that tightly grouped darts and dark darts or components can create detection challenges. That is a retailer relaying a manufacturer claim, not independent validation, and it does not establish a universal accuracy rate across configurations.
No published accuracy figure exists for LENS on a phone camera in the material supplied here. Manual scoring sidesteps recognition problems and introduces its own: entering 26 as 25, selecting the wrong segment or miscounting a visit corrupts the data exactly as a misread does.
Checking Your Number Against a Hand Tally
You do not need to audit the platform. Compare it with an independent record across several ordinary, completed 501 legs:
- Pick several complete legs, not your best match.
- Keep an independent log of each visit and, where practical, the darts thrown.
- Confirm each entered or detected score matches what landed on the board.
- Resolve visible discrepancies during play. The public material does not confirm that later corrections update history.
- Calculate the conventional average: points ÷ darts × 3.
- Compare like with like: the same player, legs and completed match Autodarts is displaying.
Include the poor scoring visits, missed doubles and weak legs. Removing them yields a highlight reel, not a match average. The test is whether the system’s figure agrees with an honest record of representative sessions.
A close match is evidence that those sessions were recorded and calculated consistently. It is not a promise about future darts, particularly after you move the camera, change the lighting, swap boards or darts, or switch scoring routes.
Use several complete matches before drawing conclusions about your level. A single result can be skewed by a fast leg, a long stint on doubles or an unusually poor scoring run.
If you suspect detection is distorting the data, compare a handful of manually entered matches with a handful of automatically scored ones. A persistent gap deserves a closer look; a small one may just be session-to-session variance.
Free Tier, Plus and X: What Paying Does and Does Not Change
Autodarts advertises core averages and match history in its free plan. Manual scoring creates recorded matches with no hardware, and the platform page describes the free tier as including core statistics and five LENS matches per month.
Autodarts Plus adds advanced statistics and the AI-referee dispute review on Autodarts X; Autodarts X itself is the £449 hardware. Paying for either does not make the underlying average more accurate. Reliability still turns on which darts are included, whether their scores were recorded correctly and how the chosen statistic is calculated.
Questions the Product Pages Do Not Answer
The documentation supplied here leaves these open:
- How early busts and winning visits finished in fewer than three darts are counted in each mode.
- Whether practice, bot, online, local and tournament games feed the same historical figure.
- How incomplete or abandoned matches are treated.
- Whether manual corrections or reviewed camera reads update historical statistics.
- Whether averages can be filtered by period, opponent, format or scoring route.
- Whether a given dashboard number covers one match, a selected period or the whole account.
These are gaps, not evidence that any game type is excluded. Read the label and the selected match-history context on screen rather than assuming a figure is a lifetime number.
Read the Checkout Percentage Beside the Average
Two players can post the same scoring average and win very different numbers of legs. Autodarts tracks checkout performance, so read it alongside the average when judging how well scoring converts into finishes.
Autodarts gives you a meaningful average for the games and throws it records, especially once you verify questionable scores and judge yourself across several complete matches. It is an Autodarts-recorded average: measurable and worth tracking, but not a complete record of every dart you throw.