Why Two Ratings for the Same Performance Can Disagree

Timeform ratings and Racing Post Ratings (RPR) are separate assessments published by different organisations. Their numbers may look compatible, but there is no reliable fixed conversion between them. The safe approach is to identify the provider and figure type, compare ratings only within the same system, and then judge whether the rated performance is relevant to today’s race.

A disagreement between the two numbers does not necessarily mean that one publisher made an error. It means two independently produced figures have been attached to the same performance. Both describe form already shown rather than guaranteeing what the horse will do next.

The short answer: do not convert Timeform ratings into RPR

A Timeform rating is not directly interchangeable with an RPR. A third-party comparison makes the limited but important point that similar-looking figures from the two systems are not equivalent point for point; it supplies neither a conversion formula nor evidence of a standard numerical gap (comparison of Timeform ratings and RPR).

That gives racecard readers two rules:

• Equal numbers need not mean equal assessments. A Timeform 120 and an RPR 120 should not automatically be treated as the same level of performance.
• A numerical gap does not establish the same gap on a shared scale. If a performance receives Timeform 128 and RPR 124, subtracting the figures does not prove a four-pound disagreement.

Each number can still be useful inside its own system. A Timeform figure can be compared with an appropriate Timeform figure, while an RPR can be compared with another like-for-like RPR. The mistake is to combine the two columns as though they came from one handicapper applying one scale.

The figures should also be treated as assessments of past form, not promises. Even a correctly understood rating cannot settle whether a horse will reproduce that performance under a different pace, trip, surface, draw, weight or state of fitness.

Timeform and RPR side by side

Timeform describes its number as an assessment of the level of form shown, while the defensible description of an RPR is narrower: it is a performance figure published by Racing Post.

Question	Timeform	RPR
Publisher	Timeform	Racing Post
Basic purpose	Expresses in pounds Timeform’s assessment of form shown	A Racing Post-published performance figure
How to read the scale	Use within the relevant Timeform scale	Use against like-for-like RPRs, not as a direct Timeform equivalent
Treatment of weight	Race Pass and Racecard displays are adjusted for weight carried	The cited Racing Post result does not state how weight is treated
Key limitation	Identify the figure type and racing discipline before comparing	Do not assume Timeform’s adjustments or calculation methods apply

Timeform says its ratings express in pounds the level of form a horse has shown. Its explanation also makes clear that a high rating can be supported by one exceptional performance; it is not automatically a measure of consistency, toughness or versatility (Timeform’s explanation of what its ratings represent).

There is a further distinction between an underlying performance assessment and the way a figure is presented on a racecard. Timeform says ratings displayed on its Race Passes and Racecards are already adjusted for the weight carried in the relevant race (Timeform’s ratings explanation).

That distinction matters. A rating quoted in a report about one performance may not be presented in exactly the same way as a figure displayed for a forthcoming race. Before comparing two numbers, establish whether each represents a particular run, a current assessment, a career peak or a weight-adjusted racecard display.

RPR should not be assumed to use Timeform’s treatment of weight, distance beaten or any other calculation input. The published RPRs can be read and compared as Racing Post figures, but Timeform’s description of its own process cannot be used as a description of Racing Post’s process.

Worked example: Kalpana’s 128 and 124

Kalpana’s King George performance shows why subtraction across the two systems is misleading.

Timeform raised Kalpana from 123 to 128 after the race, a five-point increase within Timeform’s own scale (report of Kalpana’s Timeform upgrade).

Racing Post published an RPR of 124 for Kalpana, 124 for runner-up Calandagan and 123 for third-placed Benvenuto Cellini (Racing Post result and RPRs).

Horse	Provider	Rating	Context
Kalpana	Timeform	128	Raised from 123
Kalpana	Racing Post	124	RPR for the King George
Calandagan	Racing Post	124	Runner-up’s RPR
Benvenuto Cellini	Racing Post	123	Third-place finisher’s RPR

The move from Timeform 123 to 128 is meaningful because it compares Kalpana’s old and new assessments from the same provider. It shows that Timeform raised its view of her demonstrated form by five points.

The comparison between Timeform 128 and RPR 124 is different:

128 minus 124 does not prove that Timeform rated Kalpana’s performance exactly four pounds higher on a common scale.

The calculation subtracts labels from independently produced systems. Without an established conversion, the four-point numerical difference cannot be translated into an exact difference in assessed merit.

The RPR column contains a separate lesson. Kalpana beat Calandagan by one and a half lengths, yet both received an RPR of 124 (Racing Post race report). Those published figures therefore did not simply reproduce the finishing order.

That does not reveal why the two horses received equal RPRs. The result and report give the figures and finishing margin but do not provide the handicapper’s reasoning. Attributing the equality to weight, pace or any other specific adjustment would be speculation.

The defensible conclusion is narrower: a published performance figure is an assessment, not a simple ranking generated solely from finishing position. Kalpana’s example illustrates how two rating columns can differ, but it does not supply a general conversion rule.

How to read each number without mixing the scales

Start by asking what the number represents. “This horse is rated 125” is incomplete unless the context identifies the provider and figure type. It might refer to one performance, a current assessment, a career-best rating or a figure adjusted for the weight carried in a particular race.

Then make a like-for-like comparison:

• Compare a Timeform rating with the same kind of Timeform rating.
• Compare an RPR with the same kind of RPR.
• Do not rank one horse’s Timeform figure against another horse’s RPR.
• Do not compare a career peak for one horse with a single recent run for another without acknowledging the difference.

This distinction also stops peak ability from being confused with consistency. One exceptional run can support a high Timeform rating, but the number alone does not establish that the horse regularly performs at that level. Nor does it show that the horse is equally effective over different distances, on different surfaces or under different tactical conditions.

Racing disciplines must also be kept separate. A published overview of Timeform reports that its Flat, hurdle and steeplechase ratings use non-comparable scales. The same overview places 125–129 in the average Group 1 winner band for Flat horses aged three and older (overview of Timeform’s scales and rating bands).

That band provides limited context for a Timeform Flat figure such as 128. It should not be transferred to an RPR, a hurdle rating or a steeplechase rating. An identical number can have a different meaning when the discipline or provider changes.

Finally, reconnect the figure to the performance behind it. A rating earned at today’s trip and on similar going may be more informative for the immediate task than a higher figure earned under sharply different conditions. That is a practical judgment about relevance, not a promise that history will repeat itself.

A practical racecard checklist

Use this sequence whenever a card or form report includes Timeform ratings, RPRs or both:

1. Identify the provider. Read the heading, key or publisher rather than assuming every prominent rating belongs to the same system.
2. Confirm the figure type. Establish whether it represents one run, a current assessment, a career peak or a displayed weight-adjusted figure.
3. Make a like-for-like comparison. Keep each provider’s figures within its own column and compare the same type of rating where possible.
4. Find the underlying race. Note where the figure was earned, over what distance and in what class of contest.
5. Check today’s trip and going. Ask whether the horse is being set a similar physical test.
6. Consider the wider context. Pace, carried weight, fitness, draw and recent form can all affect how relevant an earlier performance is to today.
7. Treat the rating as one input. Use it to describe demonstrated ability or help form a shortlist, not as an automatic selection.

Suppose the largest recent figure was earned over a longer trip on soft ground in a strongly run race. Today’s contest is shorter, on faster ground and likely to develop at a steady pace. The earlier rating remains evidence of what the horse achieved, but the conditions now ask a different question.

The same principle applies when a horse returns from an absence, changes distance or faces a substantially different pace scenario. The task is not to alter the published rating. It is to decide how much relevance that performance deserves in the present context.

This is racecard literacy rather than betting advice. Ratings compress information about demonstrated form, but the reader must still examine the race that produced the number and the conditions under which the horse is being asked to perform again.

What this comparison cannot tell you

The published material does not establish that either Timeform or RPR is more accurate, more predictive or more profitable. Timeform’s account explains what it intends its ratings to represent, but a publisher’s description of its own system is not independent proof of superiority.

Kalpana’s 128 and 124 also cannot establish a typical gap. It is one matched performance, not a conversion sample. Another race might produce a smaller difference, a larger difference, equal numbers or a difference in the opposite direction.

A defensible comparison of accuracy would require:

• a representative set of performances rated by both providers;
• clear separation of race-specific, current, peak and adjusted figures;
• a defined question, such as explaining past performance or predicting a later result;
• an out-of-sample test not used to construct the comparison; and
• consistent treatment of racing disciplines and race conditions.

The published Racing Post result establishes the RPRs in the Kalpana example, but it does not explain the calculation behind them. Claims about RPR weight adjustments, detailed inputs, update schedules or later revisions should therefore not be inferred from the result or borrowed from Timeform’s methodology.

The simple rule is to keep Timeform ratings in the Timeform column and RPRs in the RPR column. Use each system to understand demonstrated form on its own terms, then judge whether that form is relevant to today’s conditions. Do not manufacture a conversion or mistake either assessment for a forecast.

Does a higher Timeform rating always mean a better horse than a lower RPR?

No. A higher Timeform number cannot automatically be ranked above a lower RPR because the figures come from separate systems. First compare horses within the same provider’s scale and confirm that the figures represent the same kind of assessment.

Can one exceptional run produce a high Timeform rating?

Yes. One exceptional performance can support a high Timeform rating. The figure does not by itself establish consistency, versatility or the likelihood that the horse will repeat the performance next time.

Can I compare Timeform Flat and jumps ratings directly?

No. Timeform’s Flat, hurdle and steeplechase figures use separate scales. Keep comparisons within the same racing discipline rather than assuming identical numbers are equivalent across codes.