<?xml version="1.0" encoding="UTF-8"?><rss xmlns:dc="http://purl.org/dc/elements/1.1/" xmlns:content="http://purl.org/rss/1.0/modules/content/" xmlns:atom="http://www.w3.org/2005/Atom" version="2.0" xmlns:cc="http://cyber.law.harvard.edu/rss/creativeCommonsRssModule.html">
    <channel>
        <title><![CDATA[Stories by Jay Huskell on Medium]]></title>
        <description><![CDATA[Stories by Jay Huskell on Medium]]></description>
        <link>https://medium.com/@jay.huskell?source=rss-1fee1e8a5d91------2</link>
        <image>
            <url>https://cdn-images-1.medium.com/fit/c/150/150/0*G_e7WeguBYRTvBJR</url>
            <title>Stories by Jay Huskell on Medium</title>
            <link>https://medium.com/@jay.huskell?source=rss-1fee1e8a5d91------2</link>
        </image>
        <generator>Medium</generator>
        <lastBuildDate>Sun, 22 Nov 2026 14:48:58 GMT</lastBuildDate>
        <atom:link href="https://medium.com/@jay.huskell/feed" rel="self" type="application/rss+xml"/>
        <webMaster><![CDATA[yourfriends@medium.com]]></webMaster>
        <atom:link href="http://medium.superfeedr.com/" rel="hub"/>
        <item>
            <title><![CDATA[Is this the secret to building a winning NRL team?]]></title>
            <link>https://medium.com/@jay.huskell/is-this-the-secret-to-building-a-winning-nrl-team-1959cf74f8b5?source=rss-1fee1e8a5d91------2</link>
            <guid isPermaLink="false">https://medium.com/p/1959cf74f8b5</guid>
            <category><![CDATA[nrl]]></category>
            <category><![CDATA[sports]]></category>
            <category><![CDATA[data-analytics]]></category>
            <category><![CDATA[data-science]]></category>
            <category><![CDATA[rugby-league]]></category>
            <dc:creator><![CDATA[Jay Huskell]]></dc:creator>
            <pubDate>Wed, 27 May 2026 00:11:35 GMT</pubDate>
            <atom:updated>2026-06-02T11:16:10.685Z</atom:updated>
            <content:encoded><![CDATA[<p>Finding the best way to build a winning team has been studied in depth in all sports around the globe. The NRL is an interesting league where teams can’t build through a draft process, they cannot pay transfer fees to poach talent and they are given a salary cap that they cannot overspend. Teams have the ability to develop talent through development or “junior” systems and they are able to approach free agents in the last year of their contracts.</p><p>The Wests Tigers have been a great example for a long time on what not to do as a club. The club has a long list of former players that went onto experience success at other clubs and failed to recruit star talent in free agency. Their success on the field suffered, the year is 2026, the Tigers have not made a finals series since 2010 and finished with three consecutive wooden spoons between 2022 and 2024. At the end of 2025, the majority owners of the club removed four independent directors from the board which led to the resignation of their CEO, Shane Richardson, in what was a new low for the club.</p><p>It was a bright start to 2026 for the Tigers, they were third on the competition ladder after eight rounds with five wins, their best start to a season since 2018. Injuries have slowed down their fast start, but the club appears to be trending in the right direction. The club has made successful signings in the free agent market and promoted a plethora of junior talent into their first grade team.</p><p>If we look back to when Shane Richardson was hired as CEO of the Tigers, when interviewed on <a href="https://x.com/NBWT__/status/1734482682380472583">James Graham’s podcast ‘The Bye Round’</a>, Richardson placed great emphasis on prioritising junior development and was unimpressed with the Tigers’ recruitment policies. According to Richardson, prioritising juniors improves club loyalty and future retention, and is far more cost effective than signing from the open market. Shane Richardson was the CEO at the Penrith Panthers for their 2003 grand final win and was a long serving CEO for the South Sydney Rabbitohs who oversaw their 2014 grand final win, breaking a 43 year grand final drought, the man knows a thing or two about overseeing successful teams. So is that the secret? Is it just about bringing through talented juniors?</p><p>On first glance, this might be the case. Over the last decade, three clubs have been synonymous with success in the NRL. The Penrith Panthers won four straight grand finals and are undoubtedly the greatest modern dynasty in the NRL. They experienced noticeable player turnover in this period as salary cap pressures forced key members out of the squad, but the depth of their junior ranks ensured there were always capable replacements. The Melbourne Storm have won two premierships in that time, competed in five and are always a threat to win the competition (maybe not in 2026). The club has developed elite talent over the years with all time greats such as Cameron Smith, Cooper Cronk and Billy Slater making their debuts at the club. The Sydney Roosters won back to back premierships in 2018 and 2019, becoming the first club to do so since 1993. Whilst they have always been known for their big spending in the free agent market, the club had a host of talent who had made their debuts at the club such as Boyd Cordner, Latrell Mitchell and Joseph Manu.</p><p><strong>What is a ‘junior?’</strong></p><p>This is a topic that will never please all sides. Some fans will be appalled that the Roosters and Storm have had their junior development compared to the likes of the Panthers. The development of local juniors is the pinnacle of junior development for the loudest group of fans, so clubs with small catchment areas are heavily looked down upon for having to look outside of their own backyard for junior talent.</p><p>For data collection purposes, we will be crediting NRL clubs with developing a player if they made their debut for the club. This is a fair way to measure development as a debutant would have spent at least some time in a clubs system before making their debut except for a few fringe cases. It also would be difficult and time consuming to collect and verify each player’s junior club/district. We would also run into issues with players from rural areas who don’t come from any team’s junior league, and would we credit every New Zealand born player as a Warriors junior, regardless of whether they have had anything to do with their development system or not?</p><p>We went back and collected data from all NRL players who debuted from 2010 and onwards to capture their debut club and career games</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*7HXI-640Ce2mThm348l4Ag.jpeg" /></figure><p>It appears as though the Tigers never had a problem with developing juniors, they have debuted more players than any other club since 2010. There are strong showings from the Sydney Roosters and the Penrith Panthers who are both in the top four for most debutants since 2010. The Newcastle Knights and New Zealand Warriors round out the top five, two further clubs who have not experienced any success in this time frame. The data is underwhelming to say the least, on the eye test it is hard to say there is any correlation between developing juniors and success. But what this data doesn’t capture, is the quality of juniors being developed.</p><p><strong>What is a ‘quality’ junior?</strong></p><p>Quality is hard to quantify and can be very subjective. Since 2010 the average NRL career has lasted 69 games, with a median figure of 43. There is a large gap between the average career games and median number of career games played. Normally with the average being skewed so heavily to the right, it would be fairer to assess a quality career off the median number. But 43 games is under two full seasons played, it would be tough to argue a player had a quality career with less than two seasons under their belt. And you could argue that the right side of the skew is where the quality players are, players with lengthy careers are clearly skewing the average, those are the players we want to identify.</p><p>When we filter for players debuted who would play greater than 69 career games, we get the following:</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*kuubsdrBG58YkNI79sNQpw.jpeg" /></figure><p>These results are far more compelling. The Tigers still feature prominently, but if we look at the other clubs in the top six, the Brisbane Broncos, Penrith Panthers, Melbourne Storm, South Sydney Rabbitohs and Sydney Roosters are all premiership winners since 2010. 15 of the last 16 grand finals have featured at least one of these clubs, where they have won 12. Six of the last eight grand finals have seen these clubs make up both grand finalist spots and all eight grand finals have been won by four clubs (sorry South Sydney).</p><p>The Brisbane Broncos’ development of juniors is astonishing, they ranked 13th for total players debuted but rank first in quality juniors developed. If we take a look at clubs and their percentage of quality debutants to total debutants, the Broncos are a clear outlier.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*yX0IwUh63v3fq1_bPCY39w.jpeg" /></figure><p>Brisbane has reaped the benefits of being Queensland’s premier club and having a large catchment area for a long time. But their eye for quality is unmatched, nearly 60% of their debutants have gone on to play more than 69 games since 2010. The Cronulla Sharks jump into top four, they were grand final winners back in 2016 and they have averaged the fourth highest ladder position in the competition since 2010.</p><p>If we take a look at NRL clubs and plot their number of quality debutants against their average ladder position, a subtle trend emerges.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*9vnwL9xrD0Z_1bXiWNX3Bg.png" /></figure><p>There appears to be a moderate to positive correlation between bringing through quality debutants and a team’s average ladder position. The outlier again is the Tigers. Despite debuting more quality players than some of the most successful clubs in recent history, they have finished on average at 11.7 on the ladder. There’s a lot to say again about their recruitment and retention. As stated earlier, there have been a multitude of Wests Tigers players who have left the club and gone on to enjoy success, perhaps it’s the second and third contracts that are more important than the first. If we ignore the Dolphins who haven’t had enough time in the league to develop a meaningful sample size of juniors, the bottom seven clubs by average ladder position also take up seven of the bottom eight spots in our quality rate chart.</p><p>So is this definitive proof that focusing on junior development is the ticket to success in the NRL? Probably not. The Tigers have shown it’s not that simple. Are clubs successful because of their junior systems, or are their juniors successful because the club is successful? Successful clubs also have the luxury to be selective with the juniors they bring into first grade. While struggling clubs are hoping to develop talent they can build their club around, the successful teams already have that foundation and can be patient with their progress. While it may not be the ticket to success, the data does show that bringing through quality juniors is a good foundation.</p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=1959cf74f8b5" width="1" height="1" alt="">]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[Who are Australia’s most positive jockeys and trainers?]]></title>
            <link>https://medium.com/@jay.huskell/who-are-australias-most-positive-jockeys-and-trainers-6926182cc44e?source=rss-1fee1e8a5d91------2</link>
            <guid isPermaLink="false">https://medium.com/p/6926182cc44e</guid>
            <category><![CDATA[horse-racing]]></category>
            <category><![CDATA[data-analysis]]></category>
            <category><![CDATA[data-science]]></category>
            <category><![CDATA[sports]]></category>
            <category><![CDATA[racing]]></category>
            <dc:creator><![CDATA[Jay Huskell]]></dc:creator>
            <pubDate>Thu, 14 May 2026 05:19:09 GMT</pubDate>
            <atom:updated>2026-05-14T05:19:09.957Z</atom:updated>
            <content:encoded><![CDATA[<p><em>Speed maps and horse positioning are topics often debated by punters, trainers, jockeys and all other industry participants. We tried to quantify which jockeys and trainers are most positive.</em></p><p>Jockeys and trainers often come under heavy scrutiny for supposed “negative” tactics, which is code for having their horses settle to the rear of the field and giving their horses a perceived lesser winning chance. General wisdom and data says settling further forward does enhance your winning chances, so let’s take a look at which jockeys and trainers in Australia employ the most positive tactics.</p><p>First, we need to quantify horse settling positions. Using data available to us for Australian horse racing from 1/7/2020 and onwards, we were able to create a metric that measured how many runners a horse has settled in front of compared to how many runners they have competed against. For example, if runner A has competed against 100 runners and settled in front of 100, they have an early position rating of 100. If runner B has competed against 100 runners and has settled in front of 50 runners, they receive an early position rating of 50.</p><h3>The Trainers</h3><p>Trainers are easier to measure as they often have a consistent pattern in the way they like their horses to race and can dictate instructions to jockeys. We can take a look at which trainers have the highest average early position figure in Australia:</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/476/1*46uhCp6DVdTTpaxPqwFAnA.png" /></figure><p>Gai Waterhouse and Adrian Bott are clear outliers here. Not only are their horses settling nearly 16% further forward than the next most positive trainer, but they’ve done it with over 1,100 more runners too. Their tactics look even more spectacular when you plot all Australian trainers and compare it to their strike rate.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*K_EWlSbDradElqTqiviliQ.png" /></figure><p>Gai and Adrian are in a league of their own up there. We’ve highlighted some of Australia’s key trainers and how they position their horses on average. Most of the trainers highlighted have their horses settle forward of midfield, with the exceptions being Grant &amp; Alana Williams, Anthony &amp; Sam Freedman, and Australia’s premier trainer, Chris Waller.</p><p>You will also notice a weak to moderate correlation line between the two data points. Meaning there is a marginal correlation between a trainer’s average settle position and their strike rate.</p><h3>The jockeys</h3><p>Jockeys are trickier to measure in terms of positivity. As said before, jockey instructions can often be dictated to by trainers and other participants. If we take a look at the five most positive jockeys by settle position in Australia, one thing stands out.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/476/1*6NPaSAvOFMFSBHmeuVLK2g.jpeg" /></figure><p>Gai Waterhouse and Adrian Bott’s two preferred riders in Tim Clark and Adam Hyeronimus feature prominently on this list, while the remaining jockeys, except for Linda Meech, pair up often with trainers whose horses also settle in positive positions.</p><p>It would be unfair to reward jockeys for adhering to stable instructions. What should be rewarded are jockeys who consistently have their mounts settle further forward than they would on average. We can measure this by comparing each horse’s early position figure they carry into the race compared to their settle position on the day. We can then attribute this change to the jockey and measure their delta mean over time.</p><p>Here are the most positive jockeys on average using this method:</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/476/1*tapF4Jgj2wYf0WMLtReUvA.png" /></figure><p>Nikita Beriman is a clear standout here, having her mounts settle ahead of an extra +12.25% runners on average. The average race in South-East Queensland fields 9.6 runners, meaning Beriman’s mounts settle ahead of +1.18 horses on average. Adam Hyeronimus shows us it’s not just the stable instructions, he is a positive rider all around. Craig Williams and Linda Meech give a strong representation for Victoria here along with Jamie Melham down in eighth spot.</p><p>Here are the top five most positive jockeys by state:</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*ogRDDuvEC4ci65iK2KF9TQ.png" /></figure><p>New South Wales is led by Adam Hyeronimus with a comfortable margin on second, Rachel King. Interestingly, King has a strong relationship with the Waterhouse &amp; Bott and Bjorn Baker stables, both stables who prefer their horses to settle further forward. Regardless, she is still having her mounts settle in front of +5.2% more runners on average. Victoria has a strong contingent of positive jockeys, three of which featured in the top 10. Queensland have three jockeys inside the top ten, but this graph shows how much of an outlier Nikita Beriman is. Rosie Mahony and Codi Jordan are comfortably the most positive jockeys in their respective states of Western Australia and Tasmania.</p><p>Lastly, we can plot jockeys based on their positivity and measure it against their strike rates:</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*wHhljVMecJs6BvqSMzgHvQ.png" /></figure><p>As we can see, there’s a weak correlation between jockey positivity and jockey strike rates. What’s interesting is when we compare the correlation score with strike rates and average settle position.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*k0lvWcNxPyWPeMBUOF28TQ.png" /></figure><p>There is a small, but stronger correlation between average settle position and jockey strike rate. Meaning it’s more important for jockeys to be in a forward position than it is to have their mount settle more prominently than average. Albeit, only slightly more important.</p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=6926182cc44e" width="1" height="1" alt="">]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[How much are set restarts worth in the NRL in 2026?]]></title>
            <link>https://medium.com/@jay.huskell/how-much-are-set-restarts-worth-in-2026-8d93997d731d?source=rss-1fee1e8a5d91------2</link>
            <guid isPermaLink="false">https://medium.com/p/8d93997d731d</guid>
            <category><![CDATA[data-science]]></category>
            <category><![CDATA[data-visualization]]></category>
            <category><![CDATA[nrl]]></category>
            <category><![CDATA[data-analysis]]></category>
            <category><![CDATA[rugby-league]]></category>
            <dc:creator><![CDATA[Jay Huskell]]></dc:creator>
            <pubDate>Mon, 11 May 2026 00:23:28 GMT</pubDate>
            <atom:updated>2026-05-11T05:18:15.693Z</atom:updated>
            <content:encoded><![CDATA[<p><em>Set restarts have been handed out at record rates this season. We quantify exactly how much more valuable they are than a normal set.</em></p><p>For my first research article into the NRL, I wanted to take a look the value of a set restart in the NRL in 2026. Set restarts have been the topic of great discussion in 2026 as referees have been handing them out at record rates. So I wanted to explore and quantify how much a set restart was worth as opposed to a traditional or “normal” set.</p><p>First we need to take a look at a few data points to understand how many points are being scored and how many set restarts are being handed out on average in the NRL in 2026.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/489/1*Gcs6ElroAlXmqVSB5Gspdw.png" /></figure><p>We can see a sharp uptick in the amount of set restarts being handed out this season, as well as an improvement in the average points being scored per game.</p><p>Next, we can use NRL box score metrics, we can examine the amount of sets teams are working with each game.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/467/1*pdIR10wAJxDFcsIg_os0sw.png" /></figure><p>Without taking into account set restarts and penalties yet, teams have approximately 81 sets with the ball per game. This equates to a try every 9.2 sets or 0.109 tries per set. When accounting for set restarts and penalties, considering there are 18.6 total per game, we can bring our set count down to 62.2.</p><p>We also need to remove tries scored directly from set restarts and penalties. Using NRL play-by-play data, we can examine how many tries have occurred within one minute of a penalty or set restart occurring.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/463/1*I2c5EwPBbW_waolmdrS-_Q.png" /></figure><p>20.9% of total tries are scored directly from set restarts and 7.1% of tries are scored directly from penalties (1.84 tries per game directly from set restarts and 0.62 tries directly from penalties). Giving us a combined total of 28% of tries coming directly from infringements. For reference, in the 2018 and 2019 season combined (before set restarts were introduced), only 15.8% of tries were scored directly from penalties.</p><p>Using those numbers we can determine 0.206 tries are scored per set restart, or a try every 4.83 set restarts. Penalties aren’t very helpful, with only 0.064 tries per penalty, or a try every 15.6 penalties. This is to be expected as penalties are only given in a restricted area of the field or for foul play, meaning teams will not often receive penalties in attacking areas. What is interesting is if you open up the window to two minutes after a penalty, the percentage of tries scored jumps from 5.5% to 16.5%. While teams might not convert immediately, they are effective at building pressure from penalties.</p><p>Now that we have established 28% of tries are scored directly from penalties or set restarts, we can work out what the typical set is now worth. With an average of 80.8 sets per game, minus penalties and set restarts, we are left with 62.2 sets per game. And 8.78 tries scored per game, minus the 28% from penalties and set restarts, we are left with 6.32 tries per game. Giving us:</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/467/1*EUlixrTJ8Ma2A2YCaPxzsw.png" /></figure><p>When converted into points, tries are converted at 75% this season. Meaning each try has been worth +5.5 points this season. If we factor this into each set:</p><p><strong>Points per normal set = </strong>+0.56</p><p><strong>Points per set restart = </strong>+1.13</p><p>Meaning a set restart is worth over two times a normal set.</p><p>Lastly, we want to compare this across all seasons to examine whether this set restart value is unique to the 2026 season or if this has been the norm since set restarts were introduced back in 2020.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/464/1*MnntooLNtZGKtVnFE2_81w.png" /></figure><p>Interestingly, when it comes to converting set restarts in 2026, efficiency has actually dropped off from previous seasons. This is more than likely due to the fact that the zone in which set restarts are awarded was extended by 20m this season, meaning some set restarts will be handed out in yardage situations and not attacking situations.</p><p>What we’ve learned from this is either teams are better at converting set restarts or worse defending them. And that we can continue to expect high scoring shootouts and blowouts if set restarts continue to be handed out at their current rate.</p><p>Before signing off, I wanted to acknowledge that <a href="https://www.rugbyleagueeyetest.com/2026/05/04/adjusted-sets-and-the-points-value-of-a-set-restart/">The Rugby League Eye Test</a> did a similar project on this exact topic previously. If you haven’t checked their work our previously I highly encourage you do.</p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=8d93997d731d" width="1" height="1" alt="">]]></content:encoded>
        </item>
    </channel>
</rss>