Guy McCombe

Morning, Tim. Have you got any news or a best guess for where this Arsenal/Leicester fixture will get moved to? Thanks!

Hi Tim - a great read, thanks very much!

On the subject of Arsenal vs Leicester: do you have any info on when the reverse fixture is lining up to be rescheduled due to the clash with the FIFA champions cup? Cheers!

5 takeaways from the Women's Euro group stages The Women’s Euros kicked off at the start of July and Catalina Bush and I launched an online app using the Value Actions by Estimating Probabilities (VAEP) framework. The app is being updated after ev...

Your fave women’s footie analytics gals (@cata-bush.bsky.social and myself, ofc) are at it again!
Below is a VAEP inspired breakdown of the Women’s Euros group stages - read to see which players and teams have shown out or bombed (whoops, Netherlands 👀):
open.substack.com/pub/beyondth...

I like Zen: zen-browser.app

🤣🤣

Do you think Lyon might view her primarily as a long-term project to mold into that kind of box-crashing 10, perhaps to complement or eventually succeed Heaps? She's still young and might be worth a gamble on as an understudy, especially as a free?

Depends on how your data is stored (and what data you're using) but I like spinning up a metabase local open source environment for poking around my databases

watchin soccer

Hi Rouven, thanks for sharing it was a very interesting read. I'd be interesting in looking to extend your ANS model to incorporate it into my WSL model if you'd please be willing to share the code?

Oh so I'm way off with hopes of a surprise start tomorrow? 😅

Thanks, Tim!

Disappointed to see no official word on Russo & Kelly - do you have any info / best guess on when we might expect to see them next?

Yeah that's in there. I reckon the issue is probably how I'm handling quarter and whole goal spreads and totals. Thanks very much for your help!

To clarify: I'm optimizing to find λH and λA such that:

1) 1-Poisson.cdf(total, λH+λA) = over probability
2) Skellam.cdf(handicap, λA, λH) = probability of home covering

Yeah I think that's pretty much I what I had. Thanks for confirming - back to the drawing board for me I suppose!

That's how it's done 👏

Hi Nils, please can I ask how you derive your goal estimates from the handicap and goal lines? I'm trying to get it by inverting the poisson and skellam distributions but my subsequent models don't seem to fit very well.

Hear, hear!

Gameweek 32
We cross 1 mil together
Hand in hand, we rise

10's a bit high. I don't have the data in front of me but I think there's often a 7-8 pointer up for grabs

Box plot visualization showing top 5 Assistant Manager projections for GW31-33. Each gameweek shows 5 teams with their expected points distributions (5-95th and 25-75th percentile ranges plus mean values). GW31 leaders: Brentford and West Ham. GW32 leader: Crystal Palace and Newcastle. GW33 leader: Arsenal. Based on 100k simulations using Elevenify team ratings.

📊 GW31-33: Top Assistant Manager Projections:

GW31: Brentford & West Ham lead (mean ~9.5)
GW32: Palace and Newcastle top the chart (~13 & 12.5 resp.)
GW33: Arsenal are clear (~15.5 points)

Data: 100k sims based on @elevenify.com team ratings

Best guesses for where the champions league clashing WSL fixtures will be rearranged to?

Is McCabe suspended for Palace for a 5th yellow or does whatever happened against Chelsea not count as a yellow? Cheers 😃

this is unusual 😂 my friend found your Transfer Flow newsletter and she keepss telling me about your Variance Betting returns. she's too worried to dm u so pls can you message her? don't tell her I sent this or she will kill me lmao 😭 she really wants to know how to break into football analytics

Not dating app related but perhaps something you might want to cover - how can I share this podcast series with my friends to attempt to improve our relationship (like in ep.0) or to wean them off the redpill pipeline and improve their dating without saying: "I think you suck and this will fix you"

No of course there isn't 😂 You're welcome to lift my table calculation part from my simulator if what you're doing is python based. Just DM and I'll point you to it ☺️

I just calculate it 🙃

How can I get involved? 😁

Claude should embrace the variance and take a risk for once!

Nothing else from me, they distribute the same as Palace but slightly worse I think

Ah yes I get you now. I think all numbers within the bars are technically possible but certainly not equally likely and are distributed like your numbers there