Cover image courtesy of The Image Is Everything
There were some impressive performances posted over the weekend, but one set of times and benchmarks only tells one story. To see the full picture, you need all of the data.
Warwoven: The one to beat?
Warwoven (Sword Of State) darted away with the G1 Caulfield Guineas in good time. But was it quick enough to make him the horse to beat in the G1 Cox Plate?
Cox Plate history is littered with elite 3-year-olds taking on Australasia’s weight-for-age championship. Since 2000, only three have won: Savabeel (2004), So You Think (NZ) (2009) and Shamus Award (2013). The last two came through the Guineas—and neither won their lead-up.
Anamoe and All Too Hard won the Guineas before placing in the Cox Plate. Both rated around three to four lengths better than Warwoven did in Saturday’s Caulfield feature.
That doesn’t rule Warwoven out. The task depends on the opposition, and a weaker Cox Plate makes life easier for a 3-year-old.
On the TBF Ratings, I wouldd expect this year’s winner to need a figure around three lengths higher than Warwoven produced in the Guineas.
Land Legend’s (Fr) (Galileo {Ire}) presence matters. He brings tempo, and a genuinely run race is more likely to demand a strong rating to win. It is why European stables use pacemakers for their best horses: a fast tempo generally helps produce a fast overall time.
The market may see it differently.
Over the past year or so, I have noticed an increasing focus on last-start ratings adjusted for the weight a horse will carry next time. With Warwoven dropping to 49.5kg, it wouldn’t surprise me to see him start favourite.
Land Legend: How fast was fast?
Land Legend upset Sheza Alibi (Saxon Warrior {Jpn}) in the G1 Might And Power Stakes over 2000 metres. Since then, the conversation has centred on how hard he went early.
Much of that comes from databases comparing his speed with a standard or benchmark. And there’s a valuable lesson here: know what you are comparing against.
A benchmark draws on races run at a particular track and distance, often split further by class and track condition. In this case, we’re talking about a Group 1 over 2000 metres at Caulfield on a Good 3 that was bordering on a Firm 2.
How many races fit that description?
When you hear someone say a horse went “12 lengths above benchmark through the first 1200m”—a figure I might quote myself—ask how big the sample is and what sort of races produced it.
Now look at Land Legend’s raw times.
He averaged 12.00s per 200 metres before the 600-metre mark, then 12.15s per 200 metres over the final 600 metres. His closing splits were roughly one length slower per 200 metes than his early splits.
That comparison uses his actual times. No standard. No benchmark.
So why do the benchmark figures look so extreme?
The Might And Power is often a steadily run Cox Plate lead-up. Connections want their horses ready for the grand final, without leaving their race at Caulfield.
This time, it was run like the grand final itself.
Land Legend winning the G1 Might And Power Stakes | Image courtesy of The Image Is Everything
My database has Land Legend nine lengths faster than standard to the 600 metres, then seven lengths slower than standard over the final 600 metres. Those figures compare each section with its own benchmark; they don’t mean his actual speed collapsed by 16 lengths.
The raw times show a slowdown of about one length per 200 metres.
The benchmark figures can be correct, but “extremely fast early” doesn’t tell the whole story. The track was lightning quick, and comparable races have generally been run much more slowly early.
A benchmark tells you how a race compares with the sample. The raw splits tell you how the horse’s speed changed within that race. You need both to understand what happened. I see way too many people falling this mistake.
Black-type domination by the girls
Off a fast tempo, Lady Shenandoah (Snitzel) was given a peach by James McDonald saving ground before snatching victory in the last few bounds in the G1 Toorak Handicap.
Overall time was just inside standard but given the tempo, this race was only moderate.
Point Barrow (Blue Point {Ire}) relished the slow tempo because it meant she kept in touch and could sprint past them in the 1200-metre G3 Northwood Plume Stakes.
Reeling off the third fastest last 400 metres of the day shows how good this girl is with the right race shape. She gets a bit lost in a fast-run affair. Perhaps as she gets older, she learns to absorb better and still finish in the same manner?
My girl Surfin' Bird (The Autumn Sun) was great in the G3 Turf Vase over 1600 metres. Second-up off a slowly-run fresh contest, she was able to sit outside a genuine tempo and go away from them.
So long as Autumn Glow (The Autumn Sun) doesn’t go to the G1 Empire Rose Stakes, she looks promising.
Proxima Dea (Tarzino {NZ}) extended her unbeaten run to four when she took out the G2 Gloaming Stakes over 1800 metres at Rosehill.
The reason that she looked gone at the 200 metres is because the tempo was so slow, she was outsprinted. She was building through the line and this is the perfect race shape to platform off.
Satono Glow (NZ) (Satono Aladdin {Jpn}) was awesome in the G2 Roman Consul Stakes over 1200 metres. Clocking 1.09:2s with a last 600 metres in 33.5s is pretty impressive. She’s the one to beat in the G1 Coolmore Stud Stakes.
