Eels Blogs

Moneyball: The Art of Winning an Unfair Game was a 2003 book by Michael Lewis. Maybe you caught the movie version of 2011, starring Brad Pitt as a geeky data analyst (yeah, right). Moneyball tells the story of the Oakland Athletics baseball team, which was put together not by spit and heart but by data analytics. “Think for yourself along rational lines. Hypothesize, test against the evidence,” suggested Lewis, and this week we will look at some stats and test the following hypothesis. The Eels have an efficient-enough attack to win the competition, and a defensive and general kicking ability to restrict the opposition’ genuine chances, but the Eels fail key moments in games and let their hard-won advantages slip (H). Welcome to Round 10.

Teams

Sunday 7 May, Suncorp Stadium, BrisVegas, 6:25pm (AEDT). Lands of the Turrubal and Yuggera People. Referee: Chris Butler.

EELS: 1. Clinton Gutherson 2. Maika Sivo 3. Viliami Penisini 4. Bailey Simonsson 5. Haze Distant Dunster 6. Dylan Brown 7. Mitchell Moses 8. Wiremu Greig 9. Josh Hodgson 10. Junior Paulo 11. Shaun Lane 12. Andrew Lazarus Davey 13. J’maine Hopgood 14. Bryce Cartwright 15. Brendan Hands 16. Ryan Matterson 17. Makahesi Makatoa 18. Jakob Arthur 19. Sean Russell 20. Ofahiki Ogden 21. Matt Doorey 22. Jack Murchie.

Head coach: Brad radical-selections Arthur.

TITANS: 1. Jayden Campbell 2. Alofiana Khan-Pereira 3. Brian Kelly 4. Jojo Fifita 5. Phillip Sami 6. Kieran Foran 7. Tanah Boyd 8. Moeaki Fotuaika 9. Chris Randall 10. Tino Fa’asuamaleaui 11. David Fifita 12. Joe Stimson 13. Isaac Liu 14. Kruise Leeming 15. Erin Clark 16. Klese Haas 17. Jo Vuna 18. Keano Kini 19. Aaron Schoupp 20. Thomas Weaver 21. Thomas Mikaele 22. Sam McIntyre.

Head coach: Justin thin-ice Holbrook.

Notes: For the Titans, Brimson is out (Campbell starts at FB) and Foran is playing busted (surprise, surprise). For the Eels, Davey has been rushed into the run-on squad (to mark Fifita?), with Cartwright on the bench alongside a returning Matterson (Ogden back to reggies). With Simonsson reclaiming left centre, BA had to choose between relegating Russell or Dunster. Reliable witnesses report that BA called both players into an on-field meeting, but Dunster kept a noticeable distance from the action and thus BA selected him for his ability to report on what was going on from a reliable distance.

Observations from Last Week

Eels (vs Knights), 43-12 (W), 53% possession, 67% completion rate, 13 linebreaks, 61 tackle breaks, 16 offloads, 46 missed tackles, 7 ineffective tackles, 16 errors, 3 penalties conceded, 2 inside 10 meters, 1 ruck infringements, 0 sin bin.

Titans (vs Manly), 26-10 (L), 52% possession, 75% completion rate, 4 linebreaks, 28 tackle breaks, 7 offloads, 21 missed tackles, 13 ineffective tackles, 11 errors, 8 penalties conceded, 0 inside 10 meters, 4 ruck infringements, 0 sin bin.

Eels/Knights highlights HERE.

Titans/Manly highlights HERE

Being the Eels, even in a dominant and entertaining victory, they mismanaged some things. Their completion rate fell from 75% to 65% in the second half. They still missed tackles, though estimates varied (Nrl.com said 46, but Foxsports.com/Lab said 30). The Eels’ middle forwards played well: Grieg and (surprisingly) the bench (Ogden and Makatoa) all breaking the 150m+ barrier. BA’s comments in the presser were that the Eels ‘left a bit out there’, but they otherwise ‘looked the best, in terms of what we want to look like’, in attack.

The Titans are an enigma. In three of their losses the Titans have led in the first half but then conceded a raft of second half points: 18 against Cowboys, 22 against Dolphins, and 33 against Broncos. But they also trailed the Storm at half time before piling on 20 points in the second half to win. Who knows?

And now, some questions about the Eels after the Knights game. Why does Haze Dunster manage to repeatedly not get himself into a proper wing support play position on linebreaks? BA talks about the power game - run hard and offload - but does it seem the Eels looked best in attack when Brown and Moses were running straight and challenging the line? Are we sure this is not chicken and egg stuff? While BA thinks the power game brings the halves’ running game into the story, is it possible the halves’ running game creates enough uncertainty to free the power game? Was the Eels’ more up-tempo attacking play, halves on the run + some backfield ball movement, just poor Knights defence or a coaching acknowledgement that more conservative play is not necessarily the way to go?

Or, as Michael Lewis asked in Moneyball: “No matter how successful you are, change is always good. There can never be a status quo”.

Probabilistic Thinking

In Moneyball, Michael Lewis quips that “Baseball is a soap opera that lends itself to probabilistic thinking”. I say, Rugby League is not always too different.

Sometimes, uncertainty is first cab off the rank. Eels and Titans at Suncorp is a coin toss.

Other times, everything seems rosy. Holbrook is dreading this game.

 And if the Titans’ players were historians, they too would be dreading this game.

But fans won’t be dreading this game, because historically they have been fun to watch: lots of points, usually from both teams.

From a ‘season to date’ perspective, though, it’s clear that nothing is clear. The Eels sit 12^th from a 4/5 W/L record, and the Titans sit 10^th, 2 points above, from a 4/4/1 W/L/Bye record. The Eels have played 55% of their 9 games against 2022 Finals teams, while the Titans have played 37% of their 8 games against 2022 finals teams. But what about 2023, you ask? The Eels have played 67% of their games against teams currently in the Top 8, while the Titans have played 37% of their games against teams currently in the Top 8.

Overall, it is fair to say the Eels have had the tougher draw, and to emerge from all of that (entering Round 10) with a +34 differential versus the Titans’ -11 differential is a sign the Eels may have lost more than fans would have hoped but have otherwise not been playing terribly.

Based on per game statistics, the Eels have outperformed in attack and are middling for defence. The Eels are the third ranked team for points scored per game, and the eighth ranked team for points conceded per game. By contrast the Titans are the sixth ranked team for points scored per game, but the sixteenth ranked team for points conceded per game. Or represented as scores, the Eels’ average (rounding up) 25-21 for/against is an expected win by a try, whereas the Titans’ average 23-25 for/against is losing by a goal. Though these are averages. Looking at actual results, when the Titans lose it is by double digits, whereas the Eels’ losses have been by single digits (with one exception for each team).

And where might the game be decided? The Titans score most down their left side, where Fifita lurks. The Titans concede mostly on their right side. The Eels are relatively even where they concede their points, and score down the sides rather than through the middle. If the Eels can nullify the Titans’ left side attack, the Eels can win.

But stats can be fun. Yes, they can also be lies, damned lies, and statistics. A few weeks ago, I used some stats from The Rugby League Eye Test, so let’s do that again, with all due credit to the RLET. In one sense, the Eels are one of the stronger teams this season. Top left quadrant is ‘strong’ (high points scored, low points conceded), bottom left is ‘weak’. The Eels are flirting with weak defence, which we can say does not do justice to their attack.

Below is Figure 1, to illustrate.

But now we look at what we should expect, given a team’s field position. That is, the probability of scoring a try from specific positions on the field and weighting a team’s probability – their point value – according to starting possession from that location and the specific tackle number in their set of tackles. Or, put differently, the number of points a team should be scoring given their field position throughout a game.

Below is Figure 2, to illustrate.

For calibration, it is a miracle the Tigers are not at the top of the table, and a miracle the Knights are not dead last. The Tigers are failing to score the points they should given their field position, while the Knights are succeeding against all odds in defending poor field position (until the Eels game!).

Now toggle back and forth between Figure 1 and Figure 2, looking at the Eels and the Titans.

Start with the Titans. Given their field position, the Titans concede just a few more points than they should but score quite a few more than they should. The Titans’ attack is keeping them in games that their field position says they should not be in with a chance of winning.

But now look at the Eels. Given their field position, their points for/against in any game should be quite close. Stats don't (always) lie, and in nine games the margins (win or lose) have indeed been close: -4, -4, -4, +1, -8, +8, +26, -10, +31. The outlier results are the two big-margin victories, skewing the for/against to that +34 noted earlier. But in general, given their field position, the Eels are both scoring more than they should but also conceding more than they should.

Now we turn to Figure 3 below. A positive (higher) number is good, a negative (lower) number bad.

Figure 3.

The Sharks are super-efficient, scoring 61% more than they should, given their field position. Daylight is second, where we then find the Titans, scoring 40% more than they should. The Titans' attack is saving their gold coast fried bacon. If the Eels can limit or shut down the Titans' attack, victory is highly likely. Maybe that is why Davey was parachuted in to mark Fifita on the Eels’ right edge?

The Eels are also scoring 36% more than field position would suggest, so the Eels are also very efficient attackers. But now look at Figure 4, teams conceding more than field position would suggest. Figure 4 reverses the ranking from Figure 3, as now the lower (negative) number is better, and the higher (positive) number is worse.

Figure 4.

For calibration, the Tigers suck, and the Panthers rock. Look at the Titans. They allow 15% more points than field position would suggest. But now look at the Eels. The Eels allow 30% more points than should be expected given their field position.

Which all adds up to the following observations, to be found in the ‘The Bottom Line’.

The Bottom Line

1) The Titans score tries from what looks like low quality field position. The Titans are ranked second for efficiency of attack. The Eels need to be on alert for attacking raids from places you expect the Titans or any team to try to squeeze out a territory gain or defensive kick. The Titans might not bother to ‘play the long game’. Just run-and-shoot, as the yanks say in NFL?

2) The Eels score more points than field position would suggest (and, yes, also concede more points than field position would suggest). But the Eels thus rank as efficient attackers. Indeed (see Figure 3), the Eels are ranked third for efficiency of attack. Combine Points #1 and #2 and we can see we might witness a high scoring game from two teams that otherwise don’t appear well-placed on the field.

3) But now go back to Figure 2. The Eels are very well-placed for average expected points conceded, given field position and territory, at 16 points per game. Yes, the Eels concede one-third more than that, or about 4 points more than they should, but it shows the Eels are doing a good job controlling the field position of opponents. The Eels put themselves in a position to prevent scoring. Not as much as the Tigers, who should be conceding only 15.5, the best in the league, but instead actually concede about 24 points per game. Tigers suck. The Titans, on the other hand, are letting their opponents get into scoring positions and conceding just a little more than they should. Titans also suck (defensively).

4) Which means something very specific. The Eels are putting themselves in a position to hold teams to low points totals. The Eels are an efficient attacking team but need to translate their good control over field position into stopping teams scrounging points. Where is it going wrong for the Eels, in defence?

The Eels have a -26.4 average net post contact meters score. The Titans are worse: -54.5. The Dogs are diabolical: -108.9. Souths are +66, and both Sharks and … Tigers LOL … are +70. The Eels need to try their best to be like Souths, Sharks, and hell, even Tigers. Ironically, given we have used the Rugby League Eye Test stats to build our story, our eyeball test is that:

4a) The Eels get loose in and around the ruck, often leading to linebreaks on the edge, either side of the edge backrower. Maybe if the Eels can tighten that A and B defender, they will translate their good field position (Moses’ boot etc) into actually holding teams out? And our eyes tell us what some stats show too, for instance …

4b) The Eels are ranked fourth worst for missed tackles (on average), with Hodgson and Hopgoode (and Russell) in the Top 20 (worst).

4c) The Eels rank fourth worst for ineffective tackles (on average), and

4d) The Eels rank fourth worst for handling errors (on average), and

4e) The Eels rank second worst for errors (on average).

So, we know the ‘plays’ in which the Eels sacrifice their control of territory. Yet we know the Eels are an efficient attacking unit. Maybe fix 4a-e and the Eels can climb the ladder and go one better in 2023? The Eels are underdogs in 2023 by virtue of their middling start and because every talking media and coaching head says they lack an X-factor. But as Michael Lewis wrote in Moneyball:

“The pleasure of rooting for Goliath is that you can expect to win. The pleasure of rooting for David is that, while you don’t know what to expect, you stand at least a chance of being inspired.”

Go Eels.

PS: did I establish H, or do you think it is ~H