RadarFooty BackAccountRO

My teams

Pick the teams you follow. For each one's next match: the chances estimated by the model, in percent, and what happened in the last 10 matches.

Team rankings by condition

For each condition, the teams whose matches this season had it most often, for example “Both teams score”: 7 of 9 means that in 7 of the team's 9 matches both sides scored (1-1 counts, 2-0 does not). Next to each team is its league's rate, so you can see whether it stands out. The 95% interval shows how firm the percentage is on so few matches, and “✓ above league” appears only when even the lower end of the interval is above the league rate. It is what happened, not what comes next.

#TeamLeagueThis season95% intervalLeague rate
1BarçaLa Liga9 of 9 · 100%70–100%76%
2Real BetisLa Liga9 of 9 · 100%70–100%76%
3Sevilla FCLa Liga9 of 9 · 100%70–100%76%
4LazioSerie A8 of 8 · 100%68–100%78%
5Man CityPremier League8 of 8 · 100%68–100%73%
6MonzaSerie A8 of 8 · 100%68–100%78%
7AugsburgBundesliga7 of 7 · 100%65–100%75%
8AuxerreLigue 17 of 7 · 100%65–100%79%
9BremenBundesliga7 of 7 · 100%65–100%75%
10BrentfordPremier League7 of 7 · 100%65–100%73%

Current season, league matches only. Order = the lower end of the 95% interval, so a 5 of 5 does not move ahead of a 9 of 10 by luck alone. Teams with fewer than 5 matches are not shown.

Pick your teams

Each box shows at a glance how the team is doing: the last 5 results (W = win, D = draw, L = loss, most recent on the right), the points rate (how many of the possible points it took in recent matches, next to the league average), the state (how many of 3 measures are above the league average) and what happens to it most consistently. Tap a box to follow the team; all the details appear below. You can remove it any time.

How the box is calculated
  • Points rate = the points taken in the last 10 matches, divided by the possible ones (3 per match). League average = the same calculation over all the league's matches in the 365 days before.
  • The state counts 3 measures: (1) the points rate over the last 10 matches is above the league average; (2) the goal difference over the last 10 matches is positive; (3) the Elo rating rose over the last 5 matches (Elo accounts for the opponents' strength). Equal to the benchmark does not count.
  • 3 of 3 = good form · 2 = above average · 1 = below average · 0 = poor form. The state appears from 6 consecutive matches in our data.
  • Most consistent = at most 3 things from the last 20 matches that happened more often than in the league even in the cautious reading (the lower bound of the 95% interval is above the league rate). If there is none, the box says so.
  • “Consecutive” = no break longer than 180 days: for a team back in the league, the count starts from its return.
  • The state describes results so far, not what comes next and not the team's morale.

Your teams' next match

For each team you pick, everything this site has about it: position, Elo rating, season chances from the simulator, what has changed since your last visit, the next match (countdown, what its result can change, the model's chances and what happened in the last 10 matches), form, expected points, ongoing streaks and all its pages.

How to read it

What the percentages, the “≈ 1 in N” frequency and the record next to them mean. They describe the data and the model; they are not advice.

  • The percentage = the model's estimate for that event, calibrated on past seasons: of the matches given “60%”, about 6 in 10 had it.
  • “≈ 1 in N” = the same chance said as a frequency: at 25% it happens about once in 4 matches.
  • The record = how many times it happened in the teams' last 10 matches, with the 95% interval; a small sample gives a wide interval.
  • Estimates appear at most 14 days before the match and only for teams with at least 8 matches in the model window.
  • The teams you pick stay in this browser only; nothing is sent to us.

Updated: 25 Oct, 07:21 UTC