By The Numbers

Numbers on the side of those at the top of the sires’ table

Is success among Australian sires becoming more concentrated in the top end? The numbers suggest that may very well be the case.

A glimpse at the Australian sires’ table in mid-September can prove only so useful. Six weeks into the season, we are dealing with a relatively small sample size, with only two Group 1 races having been run and a spring packed full of features to follow.

However, one thing that does stand out is the volume of runners that have already raced this season by the top sires. Widden’s Zoustar (Northern Meteor) has already more than 150 Australian runners to the track in 2023-24 and he is closely followed by I Am Invincible (Invincible Spirit) and Deep Field (Northern Meteor) when it comes to total runners.

In all, 13 sires have had more than 100 runners in 2023-24, which is two more than had passed that milestone at the same point of last season and three more than had done so by September 14 in 2021-22. That is despite the number of races having not changed.

History tells us that this trend surrounding the increasing number of runners by the top sires has been ongoing for at least the past 15 years.   

In the 2010-11 season, the top 20 sires, judged on total runners, represented 15.3 per cent of the total horses active for that season. In 2022-23, that proportion for that same subset of sires had grown to 19.9 per cent.

In real terms, the number of horses by sires which feature in the top 20 for total runners has jumped by 18.2 per cent, from 4,780 in the 2010-11 season to 5,650 in 2022-23.

This data needs to be seen in the context of a declining stallion population, which across the same period has fallen 33.9 per cent from 790 to 522.

The vast majority of that change has been due to a decline in the number of stallions at the ‘bottom end’, who served small books for small commercial or hobby breeders. That would explain the percentage concentration among the top end, but not necessarily the significant rise in their total numbers.

The total active annual thoroughbred population, i.e. those who race each year in Australia, has dropped by 7.6 per cent since 2010-11, from 31,181 to 28,789, but the figure has remained steady over the past seven seasons.

However, since around 2016-17, the percentage of runners by the top 20 most active sires has jumped from 16.6 per cent to 19.9 per cent.

Total Australian runners per season

Season Overall runners Runners by Top 20 sires Percentage
2022/23 28,375 5,650 19.9%
2021/22 28,789 5,402 18.8%
2020/21 28,426 5,222 18.4%
2019/20 28,026 5,192 18.5%
2018/19 28,175 4,898 17.4%
2017/18 28,184 4,723 16.8%
2016/17 28,639 4,740 16.6%
2015/16 29,401 4,758 16.2%
2014/15 29,793 4,747 15.9%
2013/14 29,716 4,935 16.6%
2012/13 30,489 4,915 16.1%
2011/12 30,757 4,929 16.0%
2010/11 31,181 4,780 15.3%


Where the best comparisons can be made is when we filter by the total number of winners. The annual volume of individual winners in Australia, going back as far as 2005-06, has remained remarkably consistent, with an average of 11,860 per season. The only season that number has varied significantly was in 2007-08, which was severely disrupted by equine influenza.

In 2010-11, the number of winners by sires who ranked in the top 20 for that metric was 2,162, or 18.2 per cent of the overall sum of winners (11,899) for that season.

In 2022-23, the number of winners from the top 20 ‘winningmost’ sires was 2,648, or 22.3 per cent of the total winners for the season*.

In real terms, that means there were 486 more winners by the ‘top 20’ sires in 2022-23 than there were in 2010-11. That is a 22.5 per cent increase in the space of 12 seasons, despite the overall winning population remaining steady.

The number of winners from sires in this category took a significant jump from 2019-20 to 2020-21.

Total Australian winners per season

Season Overall winners Winners by Top 20 sire Percentage
2022/23 11,900* 2,648 22.3%
2021/22 11,932 2,520 21.1%
2020/21 11,855 2,537 21.4%
2019/20 11,679 2,355 20.2%
2018/19 11,951 2,387 20.0%
2017/18 11,973 2,365 19.8%
2016/17 11,936 2,311 19.4%
2015/16 11,972 2,238 18.7%
2014/15 11,930 2,206 18.5%
2013/14 11,361 2,296 20.2%
2012/13 12,134 2,291 18.9%
2011/12 11,950 2,197 18.4%
2010/11 11,899 2,162 18.2%

*Overall 2022/23 numbers are an approximation

Prior to this decade, no stallion had sired more than 200 winners in a single Australian season, but that has happened twice in the past three campaigns. I Am Invincible became the first to do it in 2020-21, setting an historic mark of 208 winners, and then repeated the feat in 2022-23, when he had 202 winners.

In that 2010-11 season which we are using as an historical benchmark, the leading stallion by winners was Commands (Danehill), with 155. There were 12 sires in that season with more than 100 winners, compared to 19 who reached that mark last season.

As you would expect, this concentration at the top end is reflected when we look at the breeding statistics for the total number of mares covered, compared to those covered by the top 20 most active stallions.

In 2010, the top 20 most active stallions made up 3,654 of the 24,304 covers for the breeding season, or 15 per cent. Fast forward to the most recent figures published by Racing Australia in the 2021 breeding season and that proportion had grown to 19.1 per cent.

The volume of total covers has fallen by 16.6 per cent to 20,275 in those 11 years, but the number of ‘top 20’ covers has increased by 5.9 per cent, to 3,869.

It is worth noting, however, that the proportion of top 20 sire covers has reduced slightly from its peak of 20.5 per cent in 2018. The era of the ‘big book’ may be coming to a close as the number of stallions who covered more than 200 mares in a season fell from 13 in 2018 to seven in 2021.

This change will take some time to be reflected in the racing statistics, but it will be interesting to see if the trend reported above continues to grow, or if it softens over the coming years.

Total Australian mare covers per year

Year Total Covers Covers by Top 20 busiest sires Percentage
2021 20,275 3,869 19.1%
2020 18,755 3,817 20.4%
2019 20,385 3,977 19.5%
2018 20,904 4,285 20.5%
2017 21,057 3,562 16.9%
2016 21,218 3,808 17.9%
2011-2015 Full data not available
2010 24,304 3,654 15.0%


The big question is whether this should be a concern to Australian breeding.

There is an argument that could be mounted which suggests the Australian thoroughbred industry has become more focused on commercial quality, which means a better quality of pedigree, and horse, from fewer commercial sires.

Certainly, having fewer sires which have more success has not had an adverse impact on the appeal or commercial viability of the Australian industry, which has gone from strength to strength over the past decade.

The counterargument is that more diversity at the top among stallions is better for both the genetic pool and the profile of horse that is being bred.

The proliferation of big books, and concentration of breeding among certain stallions, led The Jockey Club in the United States to briefly plan to implement a cap of 140 mares per stallion per season. It was introduced in May 2020, but rescinded in February 2022 due to commercial pressures.

The industry reaction to the ‘stallion cap’ indicates the resistance to regulatory intervention when it comes to breeding, and faith that the market and race results are best left to determine the best path.

The genetic diversity argument also falls flat when you look at how the sirelines of Australia’s top stallions have evolved since 2010-11.

In that season, six of the top 20 sires by winners were either sons or grandsons of Danehill (Danzig). In 2022-23, just four of the top 20 were sire descendants of Danehill, with those sires by 16 different stallions.     

Privacy Preference Center


Cookies that are primarily for advertising purposes



These are used to track user interaction and detect potential problems. These help us improve our services by providing analytical data on how users use this site.

_ga, _gid, _hjid, _hjIncludedInSample,