Six Months of CalledItMate: 75,000 Predictions, 104 Countries, and What I Got Wrong

Six months ago I wrote about why I built CalledItMate: five years of running football watchalongs on YouTube, thousands of people shouting at the screen, and that feeling of calling a goal two minutes before it happens. I wanted somewhere you could actually do that properly, with points that land the second you are proved right.

It went live on 21 March 2026. This is the honest update: the numbers, the features that have arrived since, and the things I got wrong.

The numbers

Since launch:

• 1,171 people have signed up, and 974 of them have made at least one prediction.
• More than 75,000 predictions have been made: 22,627 correct score calls, 25,006 goalscorer calls and 27,620 next-team-to-score calls.
• £1,300 in real cash prizes has been paid out to players. Nobody has ever paid a penny to play.
• Players have joined from 104 countries.
• The game now covers 155 competitions, from the Premier League and the Champions League down to leagues most apps never touch.
• The longest daily prediction streak so far is 90 days in a row.

None of that is huge. It is one person, no investors and no ad budget. But it is real, and every one of those numbers is a person who chose to come back.

The biggest surprise: who actually plays

I assumed CalledItMate would be a UK game. My YouTube audience is mostly British, and I built it around the matches we watch on a Saturday.

The data says otherwise. The countries with the most players are Uganda (209), Nigeria (201), Kenya (72), and only then the UK (62), followed by Algeria and South Africa.

Football is a global language, and a free game with real prizes and no stake turns out to travel further than anything I could have planned. Kick-off times show in your own timezone, and the fixture list goes far beyond Europe, because that is where a lot of the players are.

What has changed since March

The first version had the basics: predict the score before kick-off, call the goalscorer and the next team to score while the match is live, and climb a monthly leaderboard. Here is what has been added since.

Weekly seasons instead of monthly. A month is a long time to stay motivated if you start badly. Seasons now run weekly, so there is a cash prize up for grabs every single week and everyone gets a fresh start on Tuesday.

Fewer markets, done better. At one point there were six prediction markets. I cut back to the three people actually love: correct score, next goalscorer and next team to score. Fewer things to explain, more reasons to watch every minute.

Goalscorer points by position. Calling Haaland to score is not the same as calling a centre-back. Goalscorer picks now pay by position: 20 points for a forward, 25 for a midfielder, 45 for a defender and 75 if you call a goalkeeper. If your player goes on to score a brace your points double, and a hat-trick triples them.

Free picks. Call the first goalscorer of a match correctly and your next goalscorer pick on that match is free, and it stays free for as long as you keep calling them right.

The Weekly Boost. Once a week you can boost a single match, and your correct score and first goalscorer picks on it earn triple points. Choosing which match to back is half the fun.

Daily streaks. Make a prediction every day and your streak pays a daily bonus that climbs from +10 on day one to +50 a day from day seven, with milestone bonuses along the way, starting from day two. Keep a streak going long enough and you unlock a Streak Payday: a big one-off points award you can cash in at the start of a season.

Match leaderboards and awards. Big matches get their own leaderboard. Win one and you collect an award: a points multiplier for a match of your choice, an extra spin on the prize wheel, or a free match where your picks do not count against your daily limit. More than 1,400 awards have been handed out so far.

Fairness built in. Live picks on the goalscorer and next-team markets are held for a few seconds before they count, so nobody can sneak a pick in after a goal they have already seen. If VAR rules a goal out after points have landed, the points come off automatically, and if VAR rules one in, they go on. If your goalscorer pick is subbed off, you get a free re-pick.

Refer a mate. Share your link and your mate gets 100 points free the moment they join. When they play three days in a row, you get points too.

The app. CalledItMate is now on Google Play, built natively rather than as a wrapper, so you get push notifications when the goals go in.

Watch together. The thing that started all of this, the watchalong, is now built into the site: when I stream a match, you can watch and chat with everyone else playing, right next to your predictions.

And when there is no football on, there is a Penalty Arena mini-game to keep you busy.

What I got wrong

I thought getting people to sign up was the hard part. It is not. The hard part is the first week.

When I looked at the data properly, one pattern jumped out. Players who make five or more predictions in their first seven days are roughly ten times more likely to still be playing after that first week than players who make one or two. Almost everyone who leaves, leaves before they have found the fun.

So a lot of the recent work has been about that first week rather than about getting more signups: streak bonuses for coming back on day two and day three, and a nudge if you sign up and never make your first call.

I also underestimated where people would find me. In the last month, more new players arrived from ChatGPT than from Google. People are asking AI assistants for a free football prediction game, and more and more often the answer is CalledItMate. I did not plan that, but it has changed how I write every page on the site.

What the data says about football fans

Seventy-five thousand predictions also turn out to be a fascinating dataset about how fans think.

The clearest finding: fans predict draws about half as often as matches actually end in one. Across 22,095 score predictions on 5,603 matches, 1-1 was the single most common final score, yet fans predicted 2-1 about three times as often as 1-1.

Then came the twist. When I split the regular players into quarters by how often they called the result right, the most accurate quarter picked a draw far less often than the least accurate. The best predictors are not the ones who hedge on draws. They are the ones who back a winner and are right about it.

I have written that up properly, with the full numbers, here: Football prediction draws study. There is also a breakdown of the scorelines fans predict versus the ones that actually happen.

What is next

The goal has not changed: the best place in the world to call it live, free, with real money for the people who are best at it.

Everything here is still built by me, and the prizes still come out of my own pocket. If you watch football anyway, give it ten minutes during the next match you watch. Make five calls in your first week. I think you will be back.

Play free at calleditmate.com

Denver Jackson, founder of CalledItMate. I also make videos on YouTube as @Denveloper.