How to Choose a Speed-Figure System That Fits Your Analysis

Choose a system according to how you analyze races—not by comparing familiar-looking numbers across brands. Beyer figures through Daily Racing Form, BRIS, Equibase, TimeformUS, Ragozin, and Thoro-Graph differ in metric type, scale direction, adjustments, and presentation. Provider descriptions and user preferences do not establish a universal winner, so learn each system independently and compare candidates prospectively on the same races.

This is a North American-oriented comparison, not a worldwide catalog of speed-figure systems. Verify present-day availability, jurisdiction and track coverage, pricing, interfaces, and product relationships directly with each provider.

The short answer: choose a workflow, not a universal winner

Start by deciding what you need the figures and supporting data to do.

Primary workflow	Historically relevant products	What to examine
Conventional past performances	DRF/Beyer; BRIS; Equibase	Running lines, final-time figures, class context, statistics
Sectional pace analysis	BRIS; TimeformUS	Early and late pace, running styles, projected race shape
Database research	Equibase Stats RaceLens; DRF Formulator	Filtering, historical queries, customization, testing process
Sheet-pattern or form-cycle analysis	Ragozin; Thoro-Graph	Figure patterns, trip adjustments, improvement and regression cycles

This is a workflow map, not a performance ranking. Based on a historical comparison, DRF/Beyer suited readers of conventional past performances and offered Formulator customization; BRIS and TimeformUS supplied pace-oriented information; Equibase Stats RaceLens supported database research; and Ragozin and Thoro-Graph emphasized sheet patterns and trip-aware analysis. Those descriptions come from an America’s Best Racing overview published in 2020, not a current product audit.

Before choosing, distinguish two questions:

1. Does the system answer the analytical question you care about?
2. Can you obtain and use it effectively in the races you follow today?

The first can be explored through historically documented methodology and presentation. The second requires direct confirmation of current product details.

First separate speed, pace, class, and composite ratings

Not every prominent number in a past performance is a speed figure. Identify what each column measures before comparing systems.

A final-time speed figure is a normalized representation of overall race performance. Rather than comparing raw finishing times alone, it generally accounts for factors such as distance and whether the track was playing faster or slower than expected.

A sectional pace rating measures performance during a portion of the race. It might cover the start to an early call, the start to a later call, or the closing section. Two horses can earn similar final-time figures while reaching them through different distributions of early and late speed.

A race-quality or class rating addresses the caliber of the competition. Depending on the provider, a rating may describe the field or an individual horse’s performance against that field. It does not simply restate the finishing time.

A composite rating blends several inputs into one score. Speed and pace may be included alongside class, form, distance, weight, or other variables. A composite therefore answers a broader question and should not be treated as another name for a final-time speed figure.

Term	Core question answered
Final-time speed	How strong was the overall start-to-finish performance?
Sectional pace	How fast did the horse run during a specified portion?
Race rating	How strong was the race or field?
Class rating	How did the horse perform relative to the competition?
Composite	What does a blend of several factors indicate?
Variant	How is raw performance adjusted for track or daily conditions?
Par	What figure is expected for a specified class or condition?

Labels are not perfectly standardized. A “race rating” from one provider may not be constructed like a similarly named rating elsewhere. Read the provider’s definition before assuming that two columns measure the same thing.

Provider comparison matrix: scales, adjustments, and supporting tools

The following matrices summarize the 2020 descriptions cited above. They explain how the systems were presented at that time; they do not confirm current products, features, or availability.

Provider or product	Scale direction	Primary figure type	Historically documented approach
Beyer through DRF	Higher is better	Final-time speed	Distance-equivalent figures, class pars, separate dirt and turf variants, beaten-length deductions
BRIS	Higher is better	SPD plus E1, E2, and LP pace	Daily and track-to-track variants
Equibase	Higher is better	Speed, pace, race, and class ratings	Separate measures for performance, pace, and competitive context
TimeformUS	Higher is better	Speed figures and pace-oriented ratings	Pace and race-shape presentation
Ragozin	Lower is better	Sheet figure	Described as incorporating wind, carried weight, and ground loss
Thoro-Graph	Lower is better	Sheet figure	Described as broadly similar to Ragozin and trip-aware

A second view separates the headline figure from the surrounding analytical environment.

Provider or product	Historically described tools	Historical presentation	Workflow illustrated
Beyer/DRF	Formulator customization	Conventional past performances	Running-line analysis with customizable data
BRIS	Class and race measures, statistical indicators, Prime Power	Conventional data with multiple ratings	Final-time and sectional analysis together
Equibase	Stats RaceLens	Ratings plus database tools	Query-based research and testing
TimeformUS	Pace Projector, running styles, Early/Late Ratings, Race and Trainer Ratings	Pace visualization and color-coded fractions	Projected race-shape analysis
Ragozin	Figure patterns for form-cycle interpretation	Sheets omitting standard running lines and accompanying statistics	Pattern, trip, and cycle analysis
Thoro-Graph	Condensed running lines and trainer, jockey, and breeding statistics	Sheets with supplementary context	Sheet analysis with conventional details

The historical description of the Beyer method included equivalent values across distances, expected figures or pars for class levels, separate dirt and turf variants, and beaten-length deductions. That outlines the method’s general architecture, but it is not sufficient to reconstruct a proprietary calculation.

Likewise, calling Ragozin and Thoro-Graph “trip-aware” does not mean they handle every factor identically. Systems can consider similar influences yet reach different outputs because their judgments, variants, scales, and calculations differ.

Presentation matters as much as methodology. A figure embedded in a familiar past-performance format may be more usable for one analyst than a sheet graph or database interface. That is a workflow advantage, not evidence that the underlying figure is more accurate.

BRIS worked example: why one performance can need four pace and speed numbers

BRIS illustrates why final-time speed and sectional pace should be treated separately:

• E1 measures the start to the first call: two furlongs in sprints and four furlongs in routes.
• E2 measures the start to the second call: four furlongs in sprints and six furlongs in most routes.
• LP measures the second call to the finish.
• SPD measures the entire race from start to finish.

These definitions and call distances appear in the BRIS Quick Play past-performance explanation.

Imagine two horses receiving the same SPD. Horse A could have the stronger E1 and E2 profile after setting or pressing the pace, then weaken late. Horse B could begin more slowly and produce the stronger LP. The start-to-finish performances may normalize to the same SPD, but the horses reached that result differently.

That does not make sectional pace ratings inherently more predictive than SPD. The numbers answer different questions: SPD summarizes the overall performance, while E1, E2, and LP show where within the race that performance was produced.

BRIS documents different points-per-length conventions for the two kinds of rating:

• Approximately 1.5 SPD points per length in sprints
• Approximately one SPD point per length in routes
• Two pace-rating points per length for E1, E2, and LP

BRIS also says its pace and speed ratings are adjusted for daily and track-to-track variants and describes equal sprint and route SPD values as theoretically comparable. It cautions that an individual horse may not reproduce an equivalent rating after changing distance. These are provider-defined conventions within the BRIS rating system, not cross-brand conversion formulas.

Prime Power is separate from SPD. BRIS describes Prime Power as a composite incorporating speed, class, pace, form, weight, distance, and other factors. Substituting it for SPD would change the analytical question from start-to-finish performance to a blend of multiple considerations.

The points-per-length rules help interpret differences between BRIS ratings. They do not translate BRIS scores into Beyer, Equibase, TimeformUS, Ragozin, or Thoro-Graph figures. Nor does a reported win percentage for a top-rated composite establish profitability without available odds, losses, selection frequency, data costs, return on investment, and independently assessable test conditions.

Why provider figures cannot be converted point for point

A universal conversion chart is unsupported because providers do not place different labels on one shared calculation. They use different proprietary scales, variants, adjustments, and judgments. Matching numbers from two brands therefore do not establish matching performances.

Within-system normalization is not the same as cross-brand equivalence. BRIS says equal sprint and route SPD values are theoretically comparable within BRIS. That does not mean a BRIS 85 equals a Beyer 85, an Equibase 85, or a TimeformUS 85. Even within BRIS, theoretical cross-distance comparability does not guarantee that a particular horse will repeat its rating after changing distance.

Scale direction creates another obstacle. Beyer, BRIS, Equibase, and TimeformUS were described as higher-is-better systems. Ragozin and Thoro-Graph use lower-is-better figures. A declining sheet number can represent improvement, while a declining number in a higher-is-better system ordinarily indicates the opposite.

The 2020 comparison illustrated the difference between the two sheet scales by likening a Ragozin 0 to a Thoro-Graph minus-3. That specific example shows that even two lower-is-better systems do not necessarily assign the same numerical level; it is not a standing three-point conversion rule. The same historical overview explains that providers calculate and present their figures differently.

When changing providers, translate concepts rather than points. Learn:

• Which direction the scale runs
• What constitutes an ordinary, strong, or exceptional figure within that system
• How volatile the figures tend to be
• Which adjustments the provider says it makes
• How figures behave across distance, surface, class, and race type
• Whether supplementary ratings answer different questions from the headline figure

A practical provider-selection checklist

Because the non-BRIS product details in this comparison are historical, use the following checklist as a verification process rather than as a current purchasing recommendation.

1. Define the primary task. Decide whether you need final-time speed, sectional pace, class context, database research, customizable past performances, or form-cycle and trip analysis.

2. Inspect the information surrounding the figure. Determine whether the product currently includes running lines, comments, trainer or jockey statistics, race ratings, pace measures, pedigree information, or mainly sheet patterns.

3. Confirm the scale direction. Establish whether higher or lower is better, then define every supplementary metric before comparing horses.

4. Evaluate the actual format. Dense running lines, visual pace displays, database interfaces, and sheet graphs support different analytical habits.

5. Review a current sample if one is available. Inspect the actual product rather than assuming its present format from an older description. Do not assume that a provider offers a free sample or trial.

6. Verify current coverage and operations. Check pricing, trials, tracks, jurisdictions, surfaces, race types, publication timing, revision policies, historical depth, export options, and licensing directly with the provider.

7. Calculate the full workflow cost. Include purchase price, learning time, manual transcription, historical-data needs, and any limits on storing or exporting information.

Interface preferences and community recommendations can reveal usability questions but cannot demonstrate predictive superiority. A forum discussion involving DRF, Brisnet, Formulator, and Timeform can indicate what individual users value, but those preferences are anecdotal rather than controlled comparative evidence.

How to test two providers fairly

A useful comparison begins before the results are known. Looking backward can encourage selective memory: analysts may remember races in which a preferred system appeared insightful and overlook the failures.

Use a prospective process:

1. Choose a future testing period and define eligible races in advance.
2. Use exactly the same eligible races for every provider.
3. Freeze the selection rule before checking results—for example, top final-time figure, top composite, or a specified figure-gap threshold.
4. Record only information available at the stated decision time.
5. Apply the same stake or scoring rule to every system.
6. Record scratches, missing figures, publication delays, and later revisions.
7. Do not change the rule during the test. Start a separate sample for any new rule.

A practical log can remain compact:

Field	What to record
Race context	Date, track, race, surface, distance, class
Provider output	Publication time, selection or top-ranked horse, figure gap
Decision information	Available odds, stake rule, declared pass rule
Outcome and quality	Finish, gross and net return, scratches, missing data, revisions

Evaluate distinct outcomes separately:

• Ranking accuracy asks how often a system ranks horses or the eventual winner correctly.
• Win rate asks how frequently a defined selection wins.
• Profitability depends on prices, takeout, losing selections, selection frequency, staking, and data costs.
• Calibration matters if rankings are converted into probabilities: horses assigned similar probabilities should win at corresponding rates over a sufficiently large sample.

Track sample size, return on investment, and drawdown. A high win rate alone does not establish a positive return, and provider descriptions or user testimonials do not resolve that question. The same limitations apply to claims that a paid product produces consistent winners, as the cited provider-comparison discussion does not supply common-race testing, complete odds, or return data.

Avoid dividing results into numerous categories too early. After accumulating enough observations, examine sensible segments such as:

• Dirt, turf, and synthetic races
• Sprints and routes
• Different class levels
• Lightly raced and established horses

Segmentation may reveal that a system fits one analytical setting better than another, but small subgroups can create unstable and misleading patterns.

The purpose is to evaluate information under consistent conditions, not to promise winners or recommend wagers. Choose the workflow you need, verify the current product directly, learn its scale on its own terms, and compare alternatives only under identical prospective rules.

Does Racecard Stats publish its own speed figures?

The cited Racecard Stats material does not document a proprietary speed-figure product. Its racecard-reading guide discusses form, official ratings, weights, and race conditions but provides no proprietary speed-figure methodology, scale, update process, coverage list, or validation record. On that evidence, Racecard Stats should not be included as another figure provider in this comparison.

Are speed figures the same as official ratings or class ratings?

No. A speed figure generally normalizes how fast a horse performed. A class or race-quality rating concerns the strength of the competition. An official rating is an authority’s assessment used within the relevant handicapping framework and may affect assigned weight.

Because these metrics serve different purposes, similar-looking numbers should not be substituted for one another.

Does any speed-figure provider guarantee more winners or a profit?

Provider documentation can explain what a rating is intended to measure, while user reports can describe personal preferences. Neither alone establishes that one system will produce more winners or a profit.

A meaningful performance claim would require predefined selections, the same races for every provider, information available at decision time, complete odds and costs, a substantial prospective sample, and independently assessable results. Even then, accuracy, win rate, and profitability would need to be evaluated separately.