PitchPulse

Every match has a pulse!
λ Lambdas (H/A) —/—
PP Exp. Score —
🤝 Consensus —
🔥 Best Bet —
Top EV —
Calculate a match to see the 1X2 read
λ Lambdas (H/A) —/—
PP Exp. Score —
🤝 Consensus —
🔥 Best Bet —
Top EV —

League Baseline

The league's average goals per match — converts raw form into attack and defence strength. Pick a league to auto-fill, or set the numbers yourself.

Neutral venue — no home advantage
Advanced — pick a league or override
Using defaults: Home 1.50  |  Away 1.20 (toggle Advanced to change)

Team Form

Enter each side's name, then their season-to-date MP-GF-GA (home venue for Home, away venue for Away) — copy it straight off the league standings. Optionally add each side's trimmed GF Σ / GA Σ underneath: with both, Best Bet comes from the blend of the two.

HOM
AWA

Core Team Stats Required✓ Ready

Season totals off the league table — the model's main input.

Trimmed Form Optional✓ Ready

Last 5 matches (home games for Home, away games for Away), highest and lowest dropped — enter the sum of the middle 3 (GF Σ and GA Σ). Optional: with both sides filled, it is blended with the Season Tally.

Bookmaker Odds (Optional)

Type the bookies' decimal odds and they feed the EV Finder card and Model Comparison tab automatically, de-vigged behind the scenes. Same odds as the EV Finder card's own ranking on the Verdict tab — edit here and it stays in sync. Odds only? Leave the team stats blank and tap Calculate: the Verdict tab shows a Market read (de-vigged probabilities and margin, no pick).

📊 Model Comparison

How to read this

What PP predicts for the current fixture vs. what the Market expects, read from the de-vigged odds in the EV Finder card on the Verdict tab. The Market side needs odds entered there: all three 1X2 for the score picture and 1X2 tiles.

PP
⚽ Most Likely Score
Home
– – –
Away
Top correct score probability
Expected Goals
—
Over 2.5
—
BTTS Yes
—
HT Score (Est)
—

⚽ Expected Goals

Each side's λ (expected goals this match), split against the league's typical total.

1X2

Home
– –
–
Draw
– –
–
Away
– –
–
Most likely score –
Vs league baseline –

Over / Under

Over
––
–
Under
––
–
Most likely Over score –
Vs league baseline –

Both Teams to Score

Yes
––
–
No
––
–
Most likely BTTS score –
Vs league baseline –

Top correct scores

    Market Implied
    ⚽ Bookies' Most Likely Score
    Home
    – – –
    Away
    Enter odds in the EV Finder card on the Verdict tab
    Expected Goals
    —
    Over 2.5
    —
    BTTS Yes
    —
    Overround
    —

    ⚽ Expected Goals

    The same λ split, backed out of the bookies' de-vigged odds — needs all three 1X2 odds entered in Bookmaker Odds (Inputs tab).

    1X2

    Home
    – –
    –
    Draw
    – –
    –
    Away
    – –
    –
    Most likely score –
    1X2 overround –

    Over / Under

    Over
    ––
    –
    Under
    ––
    –
    Most likely Over score –
    O/U overround –

    Both Teams to Score

    Yes
    ––
    –
    No
    ––
    –
    Most likely BTTS score –
    BTTS overround –

    Top correct scores

      Calculate probabilities on the Form tab to see your Match Verdict.

      📋 Match Log

      No matches logged yet. Run the model, then hit "Add to Match Log".

      💡 How To Use PitchPulse

      Each side's season standings, one league baseline, and PitchPulse turns it into a full set of match probabilities in one tap.

      Step 1 — League Baseline: Pick a Country and League to auto-fill the average home/away goals for that competition, or type your own. Defaults are Home 1.50 / Away 1.20 — a generic mid-table read. Switching Advanced back off restores these defaults.

      Step 2 — Team names: Name your Home and Away side.

      Step 3 — Season Tally: Copy each team's season-to-date record straight off the league standings table (home rows only for the Home side, away rows only for the Away side) — matches played, goals for, goals against — into the three MP / GF / GA boxes. Neutral Venue on? Neither side is at home, so enter away rows for both teams.

      Step 3b — Trimmed form (optional, recommended): Under the Season Tally sits a second pair of boxes, GF Σ and GA Σ. For each team take its last 5 matches (home games only for the Home side, away games only for the Away side), drop the highest and the lowest score, and add up the middle 3 — goals for into GF Σ, goals against into GA Σ. Neutral Venue on? Use each team's last 5 away matches. Fill both sides and Best Bet comes from the Season Tally and the Trimmed form together; leave them blank and the pick is the Season Tally alone.

      Step 4 — Elo rating (optional): Each team-name row has an Elo pill. Leave both blank for a form-only read, or fill in both (e.g. from clubelo.com or eloratings.net) to shift expected goals toward the stronger side (the fixture's total expected goals stays exactly as the model computed it — Elo only changes how it's split) — useful when the two teams aren't from the same league or division, so their Season Tally numbers alone don't capture the quality gap. It only takes effect once both sides have a value.

      Step 5 — Calculate: Hit "⚡ Calculate Probabilities". The Model Comparison tab's PP column fills in with an Expected Goals split, 1X2, Over/Under, BTTS, and the top correct scores; the stats strip up top shows Lambdas (H/A), PitchPulse's own top correct score (PP Exp. Score), the Consensus pick (the outcome the model and the bookmaker's de-vigged odds both back, once the odds for that market are entered), the model's Best Bet — the market with the biggest cautious edge over its price (the chip shows its cautious probability), shimmering gold whenever all 3 of the top 3 most probable correct scores agree with it — and Top EV, the single positive-EV market ranked highest once odds are entered on the Inputs tab. Nothing updates until you press Calculate: changing inputs afterwards only flags the old results as out of date until you press it again.

      Step 6 — Track it: On the Logs tab (the Logs → link at the bottom of the Verdict tab), hit "+ Add to Match Log" to save the pick, then fill in the real final score once the match is played. Signal Check and Calibration build up as your log grows. The log archives automatically each day. Open 🗄 Archive for one Pending list (matches still waiting for a score) and one Results list with a single ⬇ CSV / ⬇ JSON of every match that has a final score.

      📊 Where To Get Your Team Stats

      PitchPulse doesn't pull live data — you feed it the numbers, which keeps it fast, offline-friendly, and usable for any league on earth.

      Season MP-GF-GA — matches played, goals for, goals against

      Official league standings tables — most league sites publish separate home/away standings splits (e.g. premierleague.com, laliga.com) with MP/GF/GA columns ready to copy straight in, season-to-date.

      FlashScore.com & WhoScored.com — the standings tab usually has a Home/Away split alongside the overall table.

      FBref.com — free standings and squad stats for almost every senior league worldwide, including home/away splits.

      Trimmed GF Σ / GA Σ — goals for and against over the middle 3 of the last 5 matches

      Recent results — any league site or results page lists each team's latest matches with scores; keep home games for the Home side and away games for the Away side, take the last 5, drop the highest and lowest, then add up the middle 3.

      FlashScore.com & WhoScored.com — the team's results tab has the last matches with a Home/Away filter.

      League averages

      Use the Country/League selector on the Form tab first — it auto-fills from a built-in database. If your league isn't listed, search "[league name] goals per game average" for the current season, or just leave the defaults (1.50 home / 1.20 away) as a generic baseline.

      ⚙️ Good To Know

      Everything stays on your device

      Inputs, theme, and the Match Log are saved in your browser's local storage only. There's no backend and nothing is ever sent anywhere.

      Installing PitchPulse

      Served over HTTPS, PitchPulse installs as a standalone app via the "Install App" button (or your browser's install prompt) and keeps working offline once installed.

      Light / dark mode

      Toggle the sun/moon icon in the header any time — your choice is remembered on next visit.

      ❓ Methodology & FAQ

      What model does PitchPulse use?

      A Poisson model built on each team's season-to-date record. Each team's season MP-GF-GA — matches played, goals for, goals against (home-only for the Home side, away-only for the Away side) — is blended with a league-average prior via adaptive shrinkage, so a small early-season sample doesn't overreact to a handful of results. Those shrunk rates become attack/defence strength relative to the league baseline, which convert into expected goals (λ) for each side — from there, independent Poisson distributions build the score matrix, with an adaptive Dixon-Coles / bivariate-Poisson correlation blend applied before it's renormalized (see the next question). Every market on the Model Comparison tab's PP column is derived from that final matrix.

      What is the Dixon-Coles correction, and why is it adaptive?

      Independent Poisson treats home and away goals as statistically unrelated, but real match goals are mildly coupled by things like game state and tactics — and that shows up specifically in the low-scoring cells. The Dixon & Coles (1997) tau correction adjusts exactly those four cells (0–0, 1–0, 0–1, 1–1) before the matrix is renormalized. Rather than a single fixed literature constant, ρ here is adaptive: it scales with the league's own average goals, growing more negative for defensive/low-scoring leagues and lighter for high-scoring ones — anchored so a league at the app's default baselines (1.50/1.20) gets ρ = -0.05 (lowered from the literature-standard -0.13 in v11.0, which inflated draws by about 3 points on an average Premier League fixture; tune it with the Calibration tab once you have 60+ settled games). It's derived purely from the League Baseline inputs above (never from either team's own GF/GA form) and runs automatically in the background — there's nothing to configure.

      What's the bivariate Poisson blend, and when does it kick in?

      Dixon-Coles only touches four cells — the low-score corner it was built for. Everywhere else in the matrix, independent Poisson is still uncorrelated, which matters more as a league's scoring rate climbs and more of the probability mass sits outside that corner. A bivariate Poisson model (two independent legs plus a small shared "common-shock" term, λ3, that couples every cell) fills that gap without changing either side's expected goals — λ3 is carved out of both λHome and λAway, not added on top. Which correction actually runs is itself adaptive to League Baseline: leagues at or below the app's own reference average (≈1.34 goals) run pure Dixon-Coles, leagues at 2.0+ run pure bivariate Poisson, and anything in between blends the two score matrices on a straight linear ramp. Same as Dixon-Coles above — nothing to configure, it retunes itself off the League Baseline inputs.

      Why season-to-date instead of last 5 matches?

      An earlier version read the last 5 matches and dropped the single best and worst to keep one outlier scoreline from dominating the read — that worked, but it also capped how much real evidence the model could ever use: a team's read stayed pinned to roughly 3 match-equivalents of real data all season, no matter how many games it actually played. Season MP-GF-GA uses everything played so far instead, with adaptive shrinkage (see "How are λ calculated?" below) doing the small-sample protection early on — and that protection fades naturally as the season goes on and real MP grows, rather than staying fixed.

      How are λ (expected goals) calculated?

      HomeAttack = Home team's shrunk home goals-for ÷ league average home goals. HomeDefence = shrunk home goals-against ÷ league average away goals. Away strengths mirror this using the team's away form. Before that division, each side's raw season GF/GA rate (goals ÷ matches played) is blended with a few "average" matches worth of the league baseline — so a genuine 0 across a handful of real matches doesn't collapse to a guaranteed clean sheet, while a team with a deep season sample is barely pulled off its own real rate. That blend strength is adaptive on two axes: it scales with the league's scoring level (goals follow a Poisson distribution, so a small sample is proportionally noisier in a low-scoring context than a high-scoring one — a defensive league at 0.8 goals/game gets pulled toward the baseline more firmly, around 46% weight, than a high-scoring one at 2.5 goals/game, around 33%), and it fades as real matches played (MP) grows — a team 2 games into its season leans heavily on the baseline, the same team 30 games in barely does. λHome = HomeAttack × AwayDefence × league average home goals, and λAway = AwayAttack × HomeDefence × league average away goals. A strength of 1.00× means exactly league average; 1.25× means 25% above it.

      Where do I see each side's expected goals?

      On the Model Comparison tab, in an Expected Goals card just under each column's headline score. The PP column shows λHome vs λAway as a split bar, with the combined total set against this league's typical fixture. The Market Implied column shows the same split backed out of the bookies' de-vigged odds (it needs all three 1X2 odds entered on the Inputs tab), so the two bars can be compared side by side.

      What are the default league averages, and why do they matter?

      1.50 home goals, 1.20 away goals per match — a generic mid-table baseline. Raw goal counts only mean something relative to context: 1.8 goals a game is modest in a high-scoring league and strong in a defensive one. Pick your league from the Country/League selector to auto-fill the real numbers, or enter them yourself for the sharpest read.

      What is Best Bet, and how is it different from "highest probability"?

      Best Bet asks one question: after allowing for how unsure the model is, does any market pay more than its price? For each of the seven direct markets (Home/Draw/Away, Over/Under 2.5, BTTS Yes/No) it works out a cautious probability — the model's estimate minus 0.6 standard errors (an ~73% lower bound), because expected goals built from a handful of matches are only estimates — with a small model-error allowance added so the figure also covers the model being wrong and the pick being the best of seven markets. It then compares that cautious figure with the market's price and sizes the edge with a Kelly fraction. The eligible market with the biggest Kelly fraction wins; if none has a positive edge the answer is No Bet — shown as a card with the closest market and the odds it would need (the Min odds column), and the fixture is saved to the 🚫 No Bet view of the Match Log so you can see it when picking.

      With odds entered (Markets View), the price is the odds you typed, the Verdict card shows the break-even, your edge, EV and a suggested stake (a quarter-Kelly share of your bankroll, capped at 5%), and any axis you've fully priced has its de-vigged probability blended in at 50% — bookmakers are the strongest single predictor, so the model has to earn any disagreement. Markets with no odds can't be picked. With no odds, Best Bet runs model-only: each market is priced at a typical fixture's price in this league (the fair price plus a 5% bookmaker margin), a pick must also be more likely than not on its cautious figure, and the card tells you the minimum odds worth taking. Model-only is indicative — enter odds to turn it into a real bet.

      The Verdict card's headline is the cautious probability, with a Model → Central → Cautious ladder underneath: Model is the original score-matrix estimate, Central is that estimate calibrated and blended with the market, and Cautious is the figure used for eligibility and staking. Edge is Cautious minus the break-even.

      Once at least 60 of your logged matches have a final score, every probability is also calibrated against that history (a gentle logistic correction that only moves if your results clearly show the model is over- or under-confident, or that Over 2.5 / BTTS Yes run consistently too high or too low). The top-3 correct scores are kept as a corroborating note, not a gate.

      If either team has no season tally, Best Bet won't pick a market (it shows Not enough team data) — with a blank side the model is only the league average, so there is nothing to compare against the bookmaker's price. Enter at least one match for each side.

      What is the "cautious probability", and why is it lower than the Model figure?

      Every probability in the app rests on expected goals (λ) estimated from the matches you entered, blended with a league-average prior. A team with 8 matches played gives a much shakier λ than one with 30, so Best Bet re-runs the score matrix across a spread of plausible λ values for each side and measures how far each market's probability moves. The cautious probability is the central figure minus 0.6 standard errors (an ~73% lower bound). Thin samples and early-season fixtures widen the spread, so they need a bigger edge to qualify — no separate warning flags are needed. The breakdown table shows Model, Central and Cautious side by side, and the Verdict card's headline is the cautious figure.

      Why does an amber "Inputs changed" banner appear, and why didn't the numbers update when I typed?

      Nothing is calculated, shown or logged until you press ⚡ Calculate Probabilities. Typing stats, team names, Elo, league averages, odds or form tallies only saves them. Results from your last calculation stay on screen so you can still read them, but the banner (and the dimmed header strip) tells you they no longer match the inputs. Press Calculate to refresh everything at once. Until you do, "+ Add to Match Log" is blocked so an old result can't be logged against new inputs.

      Why does Best Bet sometimes shimmer gold?

      It shimmers when the pick is unanimous — all 3 of the top 3 correct scores agree on it. The Verdict tab always shows a confirmation badge next to the pick — 3/3 or 2/3 scorelines backing it shows green "AGREE", 1/3 shows amber "NEUTRAL", and 0/3 shows red "DIVERGE" (an explicit warning that none of the model's own top-3 scorelines land on the pick) — so you can see the strength of that corroboration at a glance even when it isn't full 3/3. Only the exact 3/3 case shimmers; 2/3 is still shown as agreement, just not the unanimous version of it.

      When a Best Bet is confirmed this way and you log it, the Match Log entry itself is marked with a gold "✨ CONFIRMED" badge and the same shimmer, so it stands out from the rest of your log.

      What is the Over 2.5 screen?

      A stricter, separate check for people who want high-hit-rate Over 2.5 picks, shown as a card on the Verdict tab (and under the No Bet card). It passes when at least three of these four hold (a fixture with Over 2.5 odds entered may miss one, and the card and log then show it as 3/4; with no odds the price isn’t checked, so all three other conditions must hold): combined expected goals of at least 2.90, a cautious Over 2.5 probability of at least 58%, at least 5 matches played on the thinner side, and — if you have entered Over 2.5 odds — a cautious edge of at least 1 point over the price. With no Over 2.5 odds it reads Passes (price not checked). It never changes Best Bet, the stake or what qualifies; passing is not a guarantee. When you log a match, the result is saved with it: passing rows get a green ⚽ OVER 2.5 ✓ pill (amber with a ? if no Over 2.5 odds were entered, so the price wasn't checked, or 3/4 if one criterion was missed); when the Best Bet pick itself is Over 2.5 and the screen passed, a ✓ also appears beside the pick in the Bet Pick column, and when the pick is any other market a small +O2.5 ✓ (or +O2.5 3/4) appears there instead, so you can spot Over 2.5 candidates whatever the pick is; the filter has three options: passed (3/4 or better), 3/4 only, and full pass only, and both the Picks signal filter and the No Bet view have an Over 2.5 screen passed filter that also shows how many of the settled ones actually went over. The pill is about the Over 2.5 market on that fixture, whatever the pick was; matches logged before v11.1 have no record. The thresholds are starting values (OVER_SCREEN in best-bet.js) — log your matches and use the Calibration tab to see whether the Overs it passes really hit.

      What is the EV Finder, and how is it different from Best Bet?

      Best Bet picks the market with the best risk-adjusted edge (using your odds when you've entered them, otherwise a model-only baseline). The EV Finder lists every market at the odds you're being offered: "is each of these a good price?" You type in the decimal odds for each of the same seven markets, and it compares them against the model's own fair probability: EV% = (fair probability × odds) − 1. Positive EV means the price is better than the model's fair value; negative means it isn't — regardless of whether that market happens to be the current Best Bet. A market can be the model's top confidence pick and still be poor value if the odds are short, and a market the model rates lower can still be +EV if the price is generous. Odds you type are saved with your other inputs and survive a refresh (↺ Clear Fields wipes them), but they're only applied when you press Calculate Probabilities.

      Bookmaker De-Vig reads those same odds a different way, running quietly in the background rather than as its own visible card: instead of comparing them to the model, it strips the bookmaker's own margin (the "overround") out of each group of odds — 1X2, Over/Under 2.5, BTTS — to reveal the bookmaker's true fair probabilities. A group needs every one of its odds filled in before it can be de-vigged exactly, since the margin can't be isolated from a partial set. Once the three 1X2 odds are in, it goes a step further and searches for the λHome/λAway pair whose own score matrix would reproduce those de-vigged prices, surfacing the bookies' own implied most-likely scorelines — the same kind of read the Verdict tab gives for the model's numbers, but built entirely from the market's, so the Model Comparison tab's "Market Implied" column can show where the model and the market disagree on the shape of the match, not just the price of one outcome. If you only type one side of Over/Under 2.5 or BTTS (say just Over, or just BTTS Yes), that price still helps the λ fit: its margin is estimated from the 1X2 margin rather than the price being ignored. Only the fit (and the scorelines built from it) uses that estimate — the Over/Under and BTTS tiles, the market notes and the Match Log MKT badge still need both sides, so type both when you want the exact de-vig.

      What is Signal Check?

      A 🟢/🟡/🔴/⚪ tier for the specific bet category behind your current Best Bet (e.g. "Home Win"), computed from your own logged history using a Wilson score interval lower bound rather than a raw hit rate — so a 2-for-2 record doesn't get overrated. It needs 5+ resolved picks in that category before it rates anything; below that it shows ⚪ UNPROVEN.

      What is Calibration (the Brier score)?

      The 📊 Calibration button opens a running Brier score (0–1 scale, lower is better) split by outcome — Home Win, Draw, Away Win, Over 2.5, Under 2.5, BTTS Yes, BTTS No — comparing your all-time average against your last 10 resolved matches for each, so you can see which specific side is improving or slipping instead of one blended number per market. Where a market has 5+ resolved picks on each side, it also splits Home Win / Draw / Away Win into an Elo ON vs. Elo OFF comparison — a higher Elo ON score means the Elo adjustment is hurting that market's calibration, a lower one means there's room to lean on it more. Two plain-language tables sit under it: for every market, the average probability the model stated against how often that market actually happened, and the same for the Best Bet picks only (so you can see whether, say, the Over 2.5 picks really hit more or less often than stated); gaps are flagged beyond about ±7pt once a row has 20+ matches. The window opens with a status block: a progress bar toward full calibration, what the fit has learned (slope and intercept), and the Brier score before and after. Calibration itself starts at 30 settled matches at about half strength and phases in to full strength at 60, so a small early fit can't swing Best Bet's picks and stakes; the Best Bet card on the Verdict tab shows, for the fixture on screen, each market's raw PP probability, the calibrated one and the shift between them, and the Calibration button carries a small badge with the count and the share applied. This is independent of Signal Check: it grades the model's raw probabilities directly, rather than just whether the top pick won.

      How does the Match Log work?

      Hit "+ Add to Match Log" on the Logs tab after calculating to save a prediction. Once the match is played, enter the real final score — it's marked ✅ HIT or ❌ MISS against the Best Bet pick, and rolls into the Hit Rate shown in the stats bar. Best Bet, the Match Log, and the Model Comparison tab all grade against the same fixed Over/Under 2.5 line, so a logged "Over 2.5" always stays comparable.

      What's the Season Tally (League/Cup tally) field?

      The MP / GF / GA boxes in the Core Team Stats section — matches played, goals for, goals against, typed straight off the league standings table (home rows for Home, away rows for Away — or away rows for both when Neutral Venue is on), season-to-date. It's the main model's only input. Each box holds up to two digits and auto-advances as soon as it's full (MP → GF → GA), and after Home's GA the cursor hops straight to Away's MP — same as the Form Check tally below it. Backspace on an empty box steps back. A blank tally just means the model falls back to the pure league baseline until you fill it in — nothing errors or needs clearing first.

      What is the ⭐ Top Pick log?

      A third view in the Match Log (Logs tab), after Picks and No Bet. Whenever a Calculate run produces a Consensus pick (the outcome the model and the bookmaker's de-vigged odds both back) of 61% or more, it is saved there automatically — whether or not Best Bet found a pick for that fixture. The % is the one the Consensus tile shows (the model/market average). Enter the final score to mark it HIT or MISS — or just enter it in the Picks / No Bet log (or the Archive) and it fills in here automatically. If a later run of the same fixture drops below 61% (or the two stop agreeing), a row with no score yet is removed; a row with a score entered is kept. It has its own CSV/JSON export and is included in the ⇄ Backup file. It doesn't feed Best Bet, stakes or calibration.

      What is the 🤝 Consensus pick in the header, and how is it different from Best Bet?

      Consensus is the outcome where the model and the bookmaker's de-vigged odds agree. For each market (1X2, Over/Under 2.5, BTTS) the model's favourite (its own probability, before any market blending) is compared with the market's favourite; if they name the same outcome and each gives it at least 40% (1X2) or 55% (Over/Under, BTTS), that outcome is a consensus. If several markets agree, the strongest one is shown, with the average of the model and market probabilities. It needs all the odds for a market entered; the tile shows "—" until then and "None" when the market was read but the two disagree everywhere.

      It is independent of Best Bet: Best Bet is an edge-and-price decision and can sit on a market the market doesn't back, while Consensus only asks whether the two sources agree. It doesn't affect Best Bet, stakes or eligibility. It is logged on every Match Log and No Bet entry (a violet 🤝 badge that turns green or red once the score is in) and exported as the consensusMarket / consensusProb / consensusModelProb / consensusMarketProb / consensusResult columns.

      What is the Methods Check, and what do the ✂️ Methods: Agree / ✂️ Methods: Diverge badges mean?

      Best Bet can now come from two methods at once. Enter the Season Tally (MP-GF-GA off the standings) and the Trimmed form (GF Σ / GA Σ: last 5 matches, highest and lowest dropped, middle 3 summed) for both sides, and Calculate runs each through the same engine, blends their expected goals 50/50 and prices that blend. The Methods Check card on the Verdict tab then asks whether each method, on its own, would back the same pick: "✅ Methods agree" when both do, "⚠ Methods diverge" when one leans elsewhere, naming what it favors instead. The gap between the two methods also widens Best Bet's uncertainty, so a pick they disagree about needs a bigger edge to qualify. If either method is missing on either side, the pick is the Season Tally alone, exactly as before, and the card just says how to add the other.

      The card never logs anything by itself. But whichever way a match DOES get logged (the ordinary "+ Add to Match Log" button), the entry is tagged with the read it had at that moment: a green "✂️ Methods: Agree" or red "✂️ Methods: Diverge" badge, a "Blend" chip, and the Trimmed read's top scoreline (teal "Trim 1–1") next to the main λ, beside the bookies' de-vigged top scoreline (gold "Mkt 2–1"). No Methods badge means there was no blend at log time. Entries logged before v21.1 keep their old "📋 Form" badges, which came from a different signal (a general last-N form read), so the two are never counted together.

      What's the little time field between the team names for, and does it change the model?

      That's the match's kickoff time — entirely optional, and it never touches the model: λ, Best Bet, the Methods Check, none of it is read from or changed by what's typed there. It exists purely for the Match Log. Once a pick carrying a kickoff time is logged, it shows up as a small "🕐 15:00" chip next to the team names, and — on the live Match Log specifically, not the Archive — entries with a kickoff time are sorted latest-first, above any entry that doesn't have one; entries without a time keep their normal most-recently-logged-first order among themselves. Handy on a day with several fixtures logged out of order, so the Match Log reads top-to-bottom with the day's next-up fixture at the top. It's saved with the rest of the fixture inputs, so "↺ Clear Fields" blanks it and "↺ Reload Fixture" on a past entry restores whatever was typed for that one (press Calculate Probabilities afterwards to see results).

      What happens to the Match Log after 24 hours?

      It archives automatically. The first time the app checks after the calendar date changes — on load, or when you come back to an open tab — everything logged "today" moves into a dated archive bucket, and the Match Log starts fresh for the new day's fixtures. Nothing is deleted: open 🗄 Archive to see every match in one list: ⏳ Pending (no final score yet, with the score box right there) and 📊 Results (type a score and the match moves into Results immediately). Results has one ⬇ CSV / ⬇ JSON covering every match with a final score, oldest first, so there is a single file to audit instead of one per day. The ⬇ CSV / ⬇ JSON buttons next to "Clear all" export everything at once — today's log plus the full archive — as one file for spreadsheet or offline analysis. The 🚫 No Bet log is not rolled over — it is one running list (newest 300 fixtures) — but it has its own ⬇ CSV / ⬇ JSON buttons in the No Bet view, and the 🗄 Archive gives its fixtures the same Pending / Results split with one CSV / JSON of those that have a final score, so you can audit No Bet decisions the way you audit picks.

      Does PitchPulse store or send my data anywhere?

      No. Everything — inputs, theme, your Match Log, and its archive — lives in your browser's local storage. There's no backend and no data ever leaves your device unless you tap an export button yourself.

      Can I install PitchPulse as an app?

      Yes — served over HTTPS, PitchPulse is an installable PWA. Use the "Install App" button in the header, or your browser's own install prompt, and it'll keep working offline afterward.

      Disclaimer

      PitchPulse produces a probability estimate from recent form — not a guarantee. It doesn't account for injuries, motivation, or opponent quality beyond the season-to-date tallies you enter (an Elo rating pair, entered on the Team Form card, can correct for a cross-league quality gap, but there's no read on things like a relegation six-pointer or a dead rubber), and Signal Check / Calibration are only as meaningful as how many matches you've logged and resolved. Please gamble responsibly. — by Victor Korir

      📬 Contact the Developer

      Questions about the model, licensing, or custom builds? Reach out directly: