Small or Far Away, Father Dougal on Cows
It appears the Cow-pocalypse continues. Not only are we short of cows, but the ones we have did not produce well. Bugger.
What can we do about it? First, in the words of one great prophet, “Don’t Panic!” Really. When you have a season’s data to look back on week one is just one week, and there is perspective. When you have just one weeks worth of data, it feels like whatever a player scores is what they will score every week, or tips us off about what they will average and that is soooo not true.
Below is a list of all the cows that made it into Cow Talk last season. (I’m pretty sure.) I put their end of season averages, their scores for their first three weeks, and the differences between their averages at the end of the first three weeks and their final season averages. The “Start” column shows about when the cows first round was. A “1” is round 1, a “2” means early in the season, and a “3” means later in the season. I note as an aside how few cows worth harvesting show up later in the season.
I highlighted in green which of the three differences shown is the smallest. Not surprisingly, the more extreme the first round score, the more weeks into the season it takes for the current average to be close to their real season average. What does that mean? It means that rookies who put up a really high or low scores in round one are due to regress towards the mean, and that a rookie’s first round score was not a good indicator of their final average, especially if it was high. In fact:
- Final average of the players who had the 5 highest R1 scores = 62.6
- Final average of the players who had the 6 middle R1 scores = 69.9
- Final average of the players who had the 5 lowest R1 scores = 62.9
Oh, and the final average of all the players = 67.0
I don’t think we can conclude that an average first round score in a better indicator of a better cow, but I sure think that first round scores have very little value in predicting final scores. What a shock, small data samples have little predictive value!
How’s about after two weeks?
- Final average of the players who had the 5 highest R1 & 2 scores = 72.2
- Final average of the players who had the 6 middle R1 & 2 scores = 65.2
- Final average of the players who had the 5 lowest R1 & 2 scores = 62.0
A bit of predictive value shows up
How’s about after three weeks?
- Final average of the players who had the 5 highest R1 & 2 scores = 68.6
- Final average of the players who had the 6 middle R1 & 2 scores = 67.1
- Final average of the players who had the 5 lowest R1 & 2 scores = 61.0
Well, a bit of value in showing the lower scorers. Saw that in R2 as well.
Is there anything the cows with the top averages ended up having in common? To me, they look mostly like players who were generally expected to do well at the beginning of the season! Does that mean don’t trade your premium rookies as well as your premiums? Maybe it does.
How about the low scorers? That does seem to matter. Two crap scores in a row is not a good sign, and (surprise!) neither is three.
And what does captain hindsight say about bringing in a cow who had great scores both week 1 and week 2?
If you realized that you should have brought in the well regarded Adams and Kerridge, then you did ok. Nothing close to those first two weeks, but ok. Worth a trade? Well, they both made good money, but not enough more than an ordinary cow to make the trade worthwhile. So, not really. Much better than an injured cow though, clearly good picks for a forced trade.
If you brought in one of the ordinary-and-not-thought-to-be-really-good Papley or Ben Kennedy, then owch and dang. I started with BenKen thanks to unexpected outs, and thought I had lucked out, and I brought in Papley because look at his first two matches! But they had bad averages after Round 2, and didn’t make the big money since their big scores came too soon to help their peak price. Both made me sad. But at least I got BenKen’s points since I started with him. Papley was a total waste.
For players with slow starts, not so good to trade ’em either. It is hard to make up the cost of a trade by switching to a better rookie. The worst starters tended to make decent money unless they got hurt, which is not something we can predict.
So what does this tell us?
It is difficult to improve on your cow situation by trading cows in and out early. High risk of wasting a trade.
Unless a player is very well regarded, and for some reason you left him out of your team, don’t go with the hot streak. Great early scores are not a sign of hidden quality. (Butler, cough cough) Two rookies tonned up unexpectedly in R1 and R2 last season, and both went on to be terrible, burning their new owners like a hot flickering orangey-yellow-reddish-thing. Players generally considered to be quality are better replacement cows then whoever has had a good start.
If you are going to spend extra trades on cows, do it by bringing in downgrade targets aggressively and getting something from who you started with, even if they never hit peak or even good value.
How does this fit with the lower scores this season? Well, I can’t say for sure, but I am inclined to think the same principles apply, even if the overall averages end up lower. If a cow is hurt or just loses their job because they are crap, well, then you need to replace them. But improving by early cow-trading looks to be difficult.
Non – Cowish bit
Before the first bounce is when you come up with your plan for the season. The rest of the season you implement that plan. But the plan you had at first bounce? That is the one you are stuck with. We do not have enough trades to change plans midstream. We can fiddle, and tweak, and injuries might “force allow” us to make some changes, but implementing an imperfect plan well is going to be better than spending trades to switch to a better, (you hope, it may not be) plan with fewer trades to implement it. That’s what I think anyways.
Non – Supercoach bit
Once a year, for those who may wonder where the Small or Far away thing comes from. https://www.youtube.com/watch?v=vh5kZ4uIUC0
Leave a comment / Scroll to bottom