Why This Is Harder Than It Should Be
The horse racing tips market is crowded, and almost every service — human tipster or AI platform — claims to be profitable. That's not useful information on its own, because it's exactly what you'd expect any service to say whether or not it's actually true. The useful question isn't "do they claim good results" but "can I verify those results independently, and do the details of how they present them hold up to scrutiny." This guide is a checklist for doing that, whichever kind of service you're looking at, including this one.
1. Is the Performance Record Actually Verifiable?
The single biggest differentiator between a trustworthy service and one that isn't is whether you can see the full, dated history of selections and results — not a curated highlight reel of big winners, but every pick, win or lose, with dates and settled prices. A service that only ever shows you screenshots of winning bets, without a running, checkable log, is showing you survivorship bias by design. Look specifically for: dated picks published before the result was known (not added retroactively), a full record including losing selections, and ideally third-party verification through an independent tracking service rather than the provider marking its own homework. On this site, that record lives on the performance page, updated automatically from settled results rather than curated after the fact — which is the standard we'd suggest holding any service to, including this one.
2. How Are Strike Rate and ROI Actually Calculated?
Strike rate (the percentage of selections that win) and ROI (return on investment, accounting for the odds at which bets were actually placed) sound like simple numbers, but they can be manipulated in ways that are easy to miss. Watch for: strike rates quoted without the corresponding stakes or odds, since a high strike rate on short-priced favourites can still be a losing long-term strategy; ROI calculated using best-available prices from any bookmaker rather than a single realistic reference price, which inflates the number beyond what most subscribers could actually achieve; and results measured over a suspiciously short or cherry-picked date range rather than a full season or year. A useful, honest ROI figure is one calculated at a single, stated reference price (often Starting Price or an early-price benchmark), over a long enough sample that a lucky short-term run isn't doing the heavy lifting. Understanding how odds and implied probability work makes it much easier to spot when a quoted ROI doesn't add up.
3. AI Model vs Human Tipster: Different Failure Modes, Not Automatically Better or Worse
An AI-driven service and a traditional human tipster fail in different ways, and it's worth knowing both before assuming either is automatically more trustworthy. A model is consistent and bias-resistant but can be blind to visual form and breaking stable news; a human tipster can read subtle in-running cues but is vulnerable to recency bias, chasing losses, and simple inconsistency in workload and effort day to day. Neither structure guarantees good results on its own — what matters in both cases is whether the specific service publishes a genuinely verifiable track record, not which category it falls into. Be equally skeptical of "the algorithm can't be wrong" marketing and "trust my 20 years of experience" marketing; ask both types of service for the same verifiable data.
4. Pricing and Incentive Structure
Check how the service actually makes money, because it shapes what it's incentivised to tell you. A flat subscription fee, independent of how you bet, is a cleaner incentive structure than a service that earns commission from a specific bookmaker for every account it refers — the latter has a financial reason to push you toward a particular bookmaker's offers regardless of whether that's genuinely your best price. Also check whether a free trial or free-pick sample reflects the same selection process as the paid product, or whether free picks are deliberately easier, high-confidence bets used to build trust before the harder, more genuinely uncertain calls arrive behind the paywall.
5. Red Flags Worth Walking Away From
A few patterns are worth treating as disqualifying rather than just a minor concern: guaranteed profit language (no legitimate service can guarantee this, because horse racing outcomes are inherently uncertain); pressure tactics around joining "before the price rises" or "before the next big winner"; an inability or refusal to show a full, dated performance history when asked directly; and marketing built entirely around a handful of big-priced past winners with no visible losing selections in between. Legitimate services, AI-driven or human, are generally comfortable showing you the losing weeks alongside the winning ones, because that's what an honest long-run record actually looks like.
A Short Checklist to Use Before Subscribing
Before paying for any tips service: ask to see a full, dated, unedited results record covering at least several months; check whether ROI is calculated at a single stated reference price rather than best-available odds after the fact; check the incentive structure behind free trials and bookmaker recommendations; and be honest with yourself about whether you're evaluating the service on its actual verifiable numbers or on how confident its marketing sounds. If you're weighing this platform specifically, our own performance page and methodology guide are built to survive exactly this kind of scrutiny — and we'd encourage you to apply the same checklist to whatever service you're comparing us against. See pricing for how the subscription itself works, or start with today's picks to see the reasoning in practice before committing to anything.