ZCode Free Picks Tested: Our Results from 766 Picks Over 3 Years

Affiliate disclosure: Some links on this page are affiliate links. If you sign up through them, GamblingSociety may earn a commission at no extra cost to you. No provider pays for, reviews or approves our content. 18+ · Betting involves risk. Only stake what you can afford to lose. Help: BeGambleAware.org

ZCode publishes free picks every day on its blog: the five bets its members discuss most. But are the ZCode free picks any good? Instead of repeating marketing claims, we collected every “Top 5 Picks” post since November 2023: 966 posts with 4,534 picks, and graded every pick we could check against the real result.

The short answer: over almost three years, the ZCode free picks we could grade finished slightly in profit, with +22.7 units from 766 picks, a return of +3.0% per bet. That is clearly better than the typical bettor. It is not proof of an edge, and we explain exactly why below.

Key findings

  • 766 picks graded (football, MLB, NHL), hit rate 55.7% at average odds of 1.86.
  • Result: +22.7 units, +3.0% per bet with a flat 1-unit stake.
  • In profit in every year (2024: +5.2%, 2025: +2.0%, 2026 so far: +2.0%) and in all three sports.
  • A bettor without any edge loses around 5% per bet to the bookmaker margin. Over the same picks, that would have been roughly −38 units.
  • The honest catch: the statistical range runs from −3.6% to +9.5%. The result is not strong enough to rule out luck, and the prices did not beat the closing line.

What are the ZCode free picks?

ZCode is a paid sports betting analytics service: our ZCode System review explains the tools and prices. Every day, its blog publishes a free post called “Top 5 Picks for …”. According to ZCode, these are the top 5 most mentioned and discussed bets from the VIP Wall, the members’ discussion board.

That matters for how you read this test: the free picks are community picks, not the output of ZCode’s models. They show what the ZCode community is betting on, and whether following that crowd would have made money.

How we tested the ZCode picks

We archived all 966 daily posts from 16 November 2023 to 23 September 2026 with their exact publication time. Then we matched each pick to the real game and graded it with fixed rules, decided before we looked at the result:

  • Sports: football (22 European leagues from football-data.co.uk), MLB and NHL, using official results. Other sports could not be graded reliably.
  • Only picks with stated odds, a clearly defined bet (for example moneyline, run line, over/under, both teams to score) and a game that could be identified without doubt.
  • Only games that started after the post went online. Picks whose game was already running were excluded.
  • 1 unit per pick, settled at the odds given in the post. If a pick listed several bets, we graded the first one.
  • Not graded: parlays, player props and vague picks. We also checked the picks without odds: their hit rate (54.9%) was close to the graded picks (56.6%), so leaving them out does not flatter the result.

We are a ZCode affiliate. That is exactly why we used fixed rules and publish the method: the numbers are the same whether they help or hurt the sale. We also checked every pick at odds of 2.60 or more and 100 random picks by hand against the final score.

Result 1: slightly in profit over three years

Cumulative profit of the ZCode free picks (units)
ZCode free picks (1 unit per pick)Bettor without an edge (−5% per bet)
Cumulative profit of ZCode free picks in units, Nov 2023 to Sep 2026 −40 −30 −20 −10 0 +10 +20 +30 Nov 2023: +1.4 units after 2 picks Dec 2023: +0.8 units after 6 picks Jan 2024: +3.5 units after 16 picks 2024 Feb 2024: +6.5 units after 23 picks Mar 2024: +9.1 units after 31 picks Apr 2024: +6.8 units after 42 picks May 2024: +6.0 units after 47 picks Jun 2024: +7.4 units after 57 picks Jul 2024: +5.6 units after 68 picks Aug 2024: +10.9 units after 109 picks Sep 2024: +11.5 units after 136 picks Oct 2024: +13.2 units after 156 picks Nov 2024: +17.9 units after 187 picks Dec 2024: +11.9 units after 220 picks Jan 2025: +12.3 units after 242 picks 2025 Feb 2025: +12.4 units after 253 picks Mar 2025: +10.7 units after 285 picks Apr 2025: +16.5 units after 332 picks May 2025: +21.4 units after 382 picks Jun 2025: +19.6 units after 413 picks Jul 2025: +19.0 units after 446 picks Aug 2025: +21.3 units after 479 picks Sep 2025: +20.8 units after 500 picks Oct 2025: +15.6 units after 537 picks Nov 2025: +10.3 units after 554 picks Dec 2025: +19.1 units after 590 picks Jan 2026: +25.4 units after 617 picks 2026 Feb 2026: +25.9 units after 632 picks Mar 2026: +25.2 units after 656 picks Apr 2026: +26.9 units after 685 picks May 2026: +23.1 units after 715 picks Jun 2026: +21.8 units after 740 picks Jul 2026: +20.9 units after 748 picks Aug 2026: +22.0 units after 765 picks Sep 2026: +22.7 units after 766 picks +22.7 −38.3
Blue: running profit from betting 1 unit on every graded pick. Orange: what a bettor without an edge would lose at about 5% per bet on the same number of bets. Hover for details. Source: ZCode blog (966 daily “Top 5 Picks” posts, Nov 2023 to Sep 2026); results from football-data.co.uk, MLB Stats API and NHL API. Analysis: GamblingSociety.
SportPicksHit rateAverage oddsProfit (units)Return per bet
All76655.7%1.86+22.7+3.0%
Football (22 leagues)25255.6%1.89+10.7+4.3%
MLB36255.2%1.85+3.1+0.9%
NHL15257.2%1.86+8.8+5.8%
Flat stake of 1 unit per pick, settled at the odds stated in the ZCode post. Source: ZCode blog (966 daily “Top 5 Picks” posts, Nov 2023 to Sep 2026); results from football-data.co.uk, MLB Stats API and NHL API. Analysis: GamblingSociety.

The line is not smooth (there were losing stretches of 5 to 10 units, for example in autumn 2025) but it ended clearly above zero. Every year was positive: 2024 +5.2% (214 picks), 2025 +2.0% (370), 2026 up to September +2.0% (176).

The result also does not depend on a few lucky longshots: without the picks at odds of 2.60 or more, it was still +3.3% over 730 picks. One weak spot: heavy favourites at odds of 1.50 or lower lost −14.1% (64 picks). A small sample, but in line with our finding that short prices carry a hidden cost.

Result 2: better than the typical bettor, but luck cannot be ruled out

Return per bet with 95% range
Result per bet, bar = 95% rangeBettor without an edge (about −5%)
Return per bet with 95% range, by sport −15% −10% −5% 0% +5% +10% +15% +20% +25% All 766 picks: +3.0% per bet, 95% range −3.6% to +9.5%, hit rate 55.7%All 766 picks+3.0% Football (252): +4.3% per bet, 95% range −6.9% to +15.7%, hit rate 55.6%Football (252)+4.3% MLB (362): +0.9% per bet, 95% range −8.6% to +10.3%, hit rate 55.2%MLB (362)+0.9% NHL (152): +5.8% per bet, 95% range −9.1% to +20.8%, hit rate 57.2%NHL (152)+5.8%
Dot: result per bet. Bar: range in which the true long-run return lies with 95% probability, based on the spread of the results. Hover for details. Source: ZCode blog (966 daily “Top 5 Picks” posts, Nov 2023 to Sep 2026); results from football-data.co.uk, MLB Stats API and NHL API. Analysis: GamblingSociety.

Compared with ordinary betting, the ZCode free picks look good. In our analysis of 161,885 football matches, betting at average odds lost 3.8% per bet on favourites and 8.7% across all prices (see why most bettors lose). A slightly positive record over 766 picks is clearly better than that.

But 766 picks is not enough to prove skill. The range in the chart shows it: the true long-run return could be anywhere between −3.6% and +9.5%. Even with no skill at all, a result this good would happen by chance in roughly one case out of five.

Result 3: the closing line test

The strongest test of a tipster is closing line value: are the odds you took better than the final price before kickoff? For 98 football picks we had Pinnacle’s closing odds for exactly the same bet.

The posted odds were better than the closing price in 52% of cases: about a coin flip. At the posted prices, the expected return measured against the fair closing price was around −4.5% per bet. In other words: the football picks did not beat the market on price. Their positive result is more likely a good run than a proven edge. For MLB and NHL, no free historic closing odds exist, so this test was not possible there.


Is ZCode worth trying?

  • The free picks are a fair, free window into the ZCode community. Over almost three years they did better than most bettors do. That is more than many tipsters can show.
  • Treat them as ideas, not as a system. Our test cannot prove an edge, and the closing-line test found none in football.
  • The paid tools are something else. ZCode’s models, the Line Reversal Tool and the Game Monitor are not part of this test. If you are curious, try them yourself before you pay: there is a free trial.
  • Measure your own results. Note your odds and the closing odds for every bet: our free Bet Tracker or a spreadsheet is enough.

What this test does not show

  • It covers the free community picks, not ZCode’s paid picks or models.
  • Only 766 of 4,534 picks could be graded: football in 22 European leagues, MLB and NHL. Picks in basketball, American football, tennis and other leagues are not included.
  • The odds are the ones quoted in the posts. You may have got better or worse prices, depending on your bookmaker and the time you bet.
  • Past results do not guarantee future results.

Frequently asked questions

Are the ZCode free picks profitable?

In our test, yes, slightly. 766 graded picks from November 2023 to September 2026 made +22.7 units at 1 unit per pick, a return of +3.0% per bet. The result is not strong enough to rule out luck.

Is ZCode legit?

ZCode is a real, long-running subscription service with public daily picks. Its free picks did better than the typical bettor over almost three years. No service can guarantee profits: be wary of any claim that it can. Our ZCode review covers the paid tools.

Where do the ZCode free picks come from?

From the ZCode blog. Each daily “Top 5 Picks” post lists the five bets most discussed by members on the ZCode VIP Wall.

How did you grade the picks?

We matched each pick to the real game, graded it at the odds stated in the post with 1 unit per pick, and only counted games that started after the post was published. Parlays, player props and picks without odds were excluded.


Related guides

About the author: GamblingSociety is an independent site run by Andreas, who has operated casino and betting guides since the mid-2000s. No bookmaker or casino owns or pays for our content. More about us.