Why most records can't be trusted
A record is only worth something if the picks were written down before the games, never changed afterward, and every miss stayed in. Most sites ask you to take all three on trust. Screenshots can be made after the fact, losing picks can be quietly deleted, and a "record" that starts the day a hot streak began proves nothing.
Four questions to ask of any picks site
- When was each pick made? Look for a timestamp before the game started, on every pick, not just the winners.
- Are the misses there? Count them. A record with almost no losses, or with gaps in the schedule, has been edited.
- Graded against what? A pick should be scored at the price it was made at. "Won 8 of 10" means little if all ten were heavy favorites.
- Can you check it without trusting them? If the only evidence is the site's own word, it isn't evidence.
How a hash chain makes a record checkable
A cryptographic hash is a short fingerprint of some text: change one character and the fingerprint changes completely. Chain them, so each call's fingerprint also covers the one before it, and changing any old call breaks every fingerprint after it.
That is how CacheOut's record works. Two hours before kickoff each call is saved (the side, the probability, the number of books and the time) and never overwritten. The database writes a SHA-256 hash of that call together with the hash of the call before it.
The verify page downloads the whole chain and recomputes every hash in your own browser, and prints the command to check any single call by hand. You don't have to trust us; you can check.
What a real record still won't tell you
A verified record shows the picks were made when the site says and that nothing was changed. It does not show that following them makes money. Picks made from fair prices should be judged on calibration (do 70% calls win about 70% of the time?) rather than on a winning streak, which chance alone produces.
Not betting advice. 21+ where permitted.