Do these 3 things before closing this tab:
1Repair Windows errors before they cause bigger problems2Fix the driver behind crashes, sound loss and screen glitches3Clear out junk files and repair common Windows errorsiTechGuides is reader-supported. When you buy through links on our site, we may earn an affiliate commission. As an Amazon Associate I earn from qualifying purchases. Learn more
A useful cold-horse radar does not label a drifting runner a value bet. It tracks odds changes, compares a clearly defined market probability with an independent model estimate, and checks whether any disagreement holds up on data the model did not use. Here, “cold horse” is an article-defined shorthand for a runner whose odds drift over a stated period—not a standardized racing term, and not proof that the runner is mispriced.
What the radar should—and should not—flag
Keep three signals separate:
- Price movement: whether odds shorten or drift, relative to a stated starting price and time window.
- Model-market divergence: whether your model estimates a runner’s chance above or below the market baseline.
- Outcome and validation: whether the estimates correspond to observed results when tested on data kept separate from model fitting and selection.
A horse can drift without being a value opportunity: the model may agree with the market, or its estimate may be poorly calibrated. A shortening price—often called a steam move—does not establish that informed money caused the move or that the horse is a good bet. Liability, liquidity, market margin, new information, or participant reaction can all affect prices. Treat movement and disagreement as prompts to investigate, not verdicts.
What one study found about steam moves
Racing-Odds Labs’ study “Do Steamers Actually Win?”, published 6 July 2026, examined UK and Irish racing from 16 March to 2 July 2026. It reports 4.2 million price points across 4,135 races and 13 bookmakers. Its analysis covered 8,630 runners that were priced by at least four books and came under starter’s orders. For each runner, it averaged bookmaker opening and final pre-race prices. The study defined a steamer as an average price shortening by at least 20%, and a heavy drifter as one lengthening by at least 20%—operational cutoffs for this study, not universal standards.
Free tools Windows power users keep installed
One-click scans. No signup required.
| Study group or measure | Reported result | How to read it |
|---|---|---|
| Runners whose average price shortened at least 20% | 24.0% observed win rate | Reported for the study’s qualifying runners that started. |
| Same steamer group, opening-price implied win rate | 12.9% | The study’s opening-odds-implied figure; not margin-adjusted. |
| Same steamer group, final-price implied win rate | 20.8% | The study’s final-odds-implied figure; not margin-adjusted. |
| Runners whose average price lengthened at least 20% | 3.4% observed win rate | Reported for the study’s heavy-drifter group. |
The reported difference between the steamers’ observed win rate and opening-odds-implied figure is an association in this sample, not evidence that the move caused the result or that the signal will persist. The implied probabilities are not adjusted for bookmaker margin; the study notes that its stable control group finished below its implied figure because of that margin. Its bookmaker averages may also differ slightly from official starting prices. The findings do not establish results for other jurisdictions, periods, markets, or a future radar.
#1 Best Overall
Why track the path instead of only the last price?
A final quote compresses a changing market into one observation. A time-stamped price path can show when a move began, whether it continued, and whether it reversed. Johnson, Jones, and Tang’s 2006 study used closing prices and time-indexed price movements in a conditional-logit model of horse-race winning probabilities. Its abstract reports that closing prices did not fully incorporate market price information, particularly information less readily discernable to participants. That supports examining movement paths; it does not validate a present-day cold-horse model or establish a profitable edge.
Overreaction is another possible explanation to test. A 2019 study on prediction-market forecasts describes how prices after a movement may fail to reflect the information that initiated it, and proposes a way to adjust market-implied forecasts. This makes reversals worth recording; it does not mean that every reversal is noise or a betting opportunity.
Rank #2
- As a day trader, you can live and work anywhere in the world. You can decide when to work and when not to work.
- You only answer to yourself. That is the life of the successful day trader. Many people aspire to it, but very few succeed. Day trading is not gambling or an online poker game.
- To be successful at day trading you need the right tools and you need to be motivated, to work hard, and to persevere.
Build a comparison that can be interpreted
For each runner and observation, preserve enough context to reconstruct what the signal meant:
What’s actually slowing this PC down?
Pick the symptom - the matching free tool is one click away.
- Identity and context: race, jurisdiction, code, grade, field, and whether the runner started.
- Market observation: bookmaker or exchange, market type, price, timestamp, and whether the figure is one quote, an average, or an exchange price.
- Movement definition: baseline price, observation window, calculation method, and threshold for shortening or drifting.
- Probability baseline: how odds are converted to implied probability and whether bookmaker margin or exchange overround is addressed.
- Model estimate: the model’s probability for the same runner and outcome, generated without using future prices or results.
Do not compare raw prices from unlike markets as if they shared the same margin or liquidity. Keep opening, closing, and official starting prices distinct rather than treating them as interchangeable. A market can be a useful information-rich benchmark while still reflecting structure and behavior: the 2023 Annual Review of Financial Economics survey “Pari-Mutuel Betting Markets: Racetracks and Lotteries Revisited” reviews efficiency, pricing, mispricing, syndicates, behavioral biases, and optimization approaches.
Rank #3
- McGraw-Hill Books
- Great one for reading
- Easy to read text
Market structure can matter within a jurisdiction, too. David Marginson’s 2013 Oxford Handbook chapter analyzed Betfair starting-price overround in 2,184 UK races from 2008–2010 and reports a positive relationship between race grade and exchange overround. A separate study record on ten seasons of UK Flat Racing describes an investigation of favourite-longshot bias. These findings make race grade and favourite-versus-longshot position useful comparison axes; they do not provide one correction formula for every market or for this radar.
Independent reader supportYour contribution helps us test, update, and keep practical guides available for everyone.How to test whether model disagreements deserve trust
- Specify the target. State the jurisdiction, race types, market, observation window, starting-price definition, inclusion rules, and what counts as a move or divergence.
- Fix the market baseline. Convert prices to probabilities using a documented method. State whether and how margin or overround is handled; if it is not handled, say so.
- Keep the model independent. Produce its probability without feeding it information that would only have been available later than the timestamp being evaluated.
- Reserve untouched evaluation data. Do not use the same races to fit or select the model and to claim its performance. Group or otherwise assess predicted probabilities against observed frequencies on data not used in those decisions.
- Report the scope with the result. Give the sample size, time period, geography, market coverage, and inclusion rules. Evaluate probability quality rather than relying only on winner-picking accuracy or short-run returns.
- Check for context and reversals. Compare results by market type, race grade, favourite-longshot position, and movement shape, and examine whether apparent disagreements persist or reverse.
Calibration is a validation requirement, not a property established for a cold-horse radar by the studies cited here. No cited result shows that a particular model’s probability estimates are calibrated or that model-market divergence predicts profit. A persuasive claim therefore needs an explicitly scoped, out-of-sample evaluation—not just a striking odds chart or a run of winners.
Quick Recap
Best Value
- Language: english
- Book - trading: technical analysis masterclass: master the financial markets
- It is made up of premium quality material.
Product prices and availability are accurate as of the date/time indicated and are subject to change. Any price and availability information displayed on Amazon at the time of purchase will apply.
