SportEZ

Week in review: 7 to 14 September. Second losing week in a row.

Record: 7 won, 8 lost. Down 4.55 units on 21 staked, which is minus 21.7% on turnover. Last week was minus 22.2%.

By competition: NPB 4W-3L, plus 0.19u. KBO 3W-5L, minus 4.74u.

Best pick: Lotte Giants to win at 1.87, KBO, 12 September. Won at 2 units, plus 1.74u.

Worst pick: LG Twins to win at 1.67, KBO, 8 September. Lost at 3 units, minus 3.00u.

The pattern worth naming: both losing weeks came from the same league. KBO has given back 6.37 units in two weeks after making 12.28 over the previous two months. We are not pausing it. Sixteen bets is nowhere near enough to call a trend, and when we paused a league in July we did it on fifty five settled bets, not sixteen. Reacting to two bad weeks is how you end up switching things off and on at random. It is on watch, and if the next fifty bets look like these we will say so here.

Context: across all 310 settled bets the hit rate is 53.2% and the yield is now slightly below zero. It was above zero a fortnight ago. That is what a real record does, it moves, and we would rather show you the down leg than wait for a good month to start posting again.

Full record, every settled bet: https://sportez.pro/record/

18+. Not betting advice. Gambling can be addictive, please play responsibly.

How a pick actually gets made

Here is the real process, including the parts we deliberately do not do.

First, a number built from results. Ratings come from every result of the season, not from opinions about form or which side looks hungrier.

Then a second, independent read. The same match is assessed again from scratch, and the two reads are kept apart so neither can talk itself into agreeing with the other. When they disagree sharply, we usually skip the match.

Then a news check before anything goes out. Lineups and injuries. A pick built on yesterday's information gets dropped rather than sent late.

Edge and stake are computed separately, in code, so nothing grades its own homework on how confident it should sound.

Hard limits kill picks: too little season history, too large a disagreement with the market, or an edge below threshold. Most matches produce nothing. On a normal day we look at twenty or more and publish three. No bet is a real answer here, not a failure to find one.

What we do not do, despite it being common in this industry: referee profiling, cards and corners markets, weather models. We do not offer those markets, so we do not pretend to analyse them.

Full write-up: https://sportez.pro/methodology/

18+. Not betting advice. Gambling can be addictive, please play responsibly.

Most tipsters run a confidence game. Here is how we are different.

The open secret of this industry: a lot of "professional tipsters" are not profitable. Wins get posted loudly, losses get deleted quietly, and the record resets whenever the numbers get uncomfortable. SportEZ was built as a direct answer to that.

Every pick is logged before kickoff. Timestamp, odds, stake, reasoning. It is settled automatically against the final score and it stays up whether it won or lost. The full record is public and you can check it yourself: https://sportez.pro/record/

We publish our own bad numbers too. Across 310 settled bets the hit rate is 53.2% and the yield is currently sitting slightly below zero. When we say a pick is about 63% likely, it lands about 53% of the time. The market prices these games better than we do. That figure moves in both directions and the public record always shows where it actually stands, not where we would like it to stand.

We act on it. In July we switched off baseball's biggest league because our own record said it was losing money there. In September we switched off football over and under 2.5 goals, after 41 of 42 such picks landed on the same side of the line. That is a lean, not an edge.

No pick on a dead day. When there is no edge in the market, nothing gets sent. Forcing action is how recreational bettors lose money.

What we actually cover: football across several European and South American leagues, Korean and Japanese baseball, and WNBA basketball. Not tennis, not NFL. We go where the market is thin, not where the audience is biggest.

How a pick gets made, in full: https://sportez.pro/methodology/

18+. Not betting advice. Gambling can be addictive, please play responsibly.