Cole O'Dell

The haters saying Oklahoma was the next Nebraska… honestly great call from the haters

@FanDuel Incorrect

The Cornhuskers took over "Loud City" 📢 Nebraska set the db record at the Paycom Center in OKC beating out the Thunder's previous record 😤

@dcurt15 @SmashAccept @DynastyDadFF Smash smash smash

@theblockspot MVP? 🤣

Jaylen Brown was decent. Then Jaylen Brown was solid. Then Jaylen Brown was good. Then Jaylen Brown was very good. Then Jaylen Brown was great. Then Jaylen Brown was a star. Then Jaylen Brown was a superstar. Jaylen Brown is now one thing only. The MVP of the whole damn league.

@MikeBellATL @fia @F1 @Shell @IndyCar wait is the battery love just for sustainability? i figured it was some engineering improvement (new fan)

The @fia has ruined @F1. You want battery management watch Formula E. You could run V-8's on the same kind of sustainable biofuel @Shell makes for @IndyCar and check the box for green agenda. Give us back pedal to the metal racing.

@TravisSkol @KevinOConnor Kevin won’t reply to this

@KevinOConnor Is this basketball? Jaylen’s coming across a little lacking in self awareness. He was hunting contact late while SGA didn’t shoot a single 4Q FT.

Jaylen Brown is right: That’s not basketball. But I can’t blame SGA for it. I blame the referees and the league itself for rewarding this behavior. It should not be a foul. Defenders should be allowed to play defense. Marginal contact like this should not draw a whistle. Play on.

@DavisMattek So funny I was vibe coding this exact thing like 2 months ago. Are you scraping values from KTC and FantasyCalc? One avenue I was trying to do was some sort of blend between calculators for a “market value” then trying to get “expert value” (rankings) to signal buys and sells

I've been working on a dynasty trade calc that attempts to solve the most obvious problem in dynasty: an "asset" or player is not worth the same to all teams I've taken market based values and applied an adjustment based on age and predicted future value that allows you to run your team as "win now", "tanking", "neutral" or "productive struggle Link down below if you want to mess around with it

@steipete @mstr_brad @openclaw Without a doubt

@mstr_brad @openclaw Cl***e is so much better?

Interesting benchmark on which model is best for @openclaw pinchbench.com

Me sitting there knowing the state of Oklahoma exists

@Geiger_Capital Anthropic has been very clear that AI is not at the level where it should be autonomously killing people. The government should understand that. This isn’t the military building a bomb, this is a private company saying their product is not capable of it.

This is moronic… Fully autonomous weapons are happening. You can’t stop it. The only thing Anthropic is doing is ensuring that China will have them and we won’t.

@PeteHegseth @AnthropicAI @DarioAmodei @POTUS Seems entirely reasonable that companies don’t want you to use their products to not spy on Americans or autonomously kill people with no human oversight

Thank you for your attention to this matter. cc: @AnthropicAI @DarioAmodei

@ffdataroma Fortunately it won’t be Justin Fields, it’ll be Malik Willis 🫩

With Breece Hall seemingly more and more likely to remain a Jet, I actually think his outlook is going to be over-scrutinized this offseason. Breece has shown an elite ceiling in fantasy, pairing explosive runs and heavy receiving work together (elite career 6.3% explosive rate and 12.1% target share). In 2025, Breece and the Jets offense was extremely hindered by QB play. Fortunately, it won’t be Justin Fields anymore, who simultaneously crushed Breece's TD upside (5 TDs, 8+ in each of last 2 seasons) and receiving volume (2.6 targets/game, 4.5+ in each of last 2 seasons). The Jets have a solid OL, and they should add a 2nd pass catcher and at least competent QB. There’s a lot of negative vibes around them because they’re "the Jets", but the offensive environment might not be as bad as many are thinking. I'm still fine with him at ADP, and any drop would be welcomed/added value IMO.

@warnerbro @ffdataroma Gibbs had a much lower snap share his rookie year than Jeanty. Gibbs also had 90 targets last season. Walker and Charb combined for 58 targets. I think it would be highly unlikely that Jeanty gets into 80+ target range, which he would need to pay off his current value.

@CptStugots @ffdataroma He had more fantasy points as a receiver than Gibbs did his rookie year in the worst offensive environment in the NFL.

This quote feels like a throwaway IMO. Basically every team gives work to two RBs, and this gives us no information on what the workload split for Ashton Jeanty will look like. We constantly see coaches adapt their usage and tendencies based on the talent on their teams. Jeanty is going to be light years more talented than any other RB LV has and/or brings in, so pattern matching to last year’s Seattle backfield (pair of round 2 RBs that have proven successful at NFL level and have differing strengths) feels misguided. We should see major environment and efficiency improvements for Jeanty this year, and hopefully more scoring opps with that. A couple less touches per game (if that even happens) is no worry.

@ffdataroma I’m concerned about Jeanty’s ceiling. Clearly talented, but I think the path of him paying off his elite fantasy draft capital just got much less likely imo. Don’t think Jeanty gets Bijan/Gibbs type receiving usage to prop him up.

@CptStugots Oh yea definitely, I just think that's rather unlikely

@andrewsurp123 @ffdataroma Bijan and Gibbs are also heavily utilized in the pass game with 90+ targets last year. Walker and Charbonnet saw a combined 58.

@ffdataroma That is what I’m thinking as a Jeanty owner. You hardly see a bell cow rb in today’s game like look at bijan and Gibbs for example they had very high calaber RBs behind them and were still able to put up elite production

@calebhodges @openclaw I always get this error when i restart the gateway HTTP 401: authentication_error: Invalid bearer token

Big news – the newest @openclaw update today quietly re-added oAuth for Anthropic Subscriptions! I just switched back to Sonnet 4.6 as my main without using the expensive API Token. Here's how in 60 seconds: Run "claude setup-token" in your terminal and save your token. Run "openclaw models auth add" Select anthropic → setup-token → paste Restart your gateway & open a new chat That's it. Your Pro/Max subscription now powers your local OpenClaw setup. No API key billing. Thanks @cathrynlavery for letting me know! @AndrewWarner We're back to Claude!

@iamsupersocks Hmm. I kept getting a forbidden error anytime I put it in

@CptStugots Openclaw configure in the terminal, that's all.

J’ai cramé 200 $ de tokens Claude en 30 minutes sur OpenClaw. Hier après-midi, je codais tranquille avec mon agent sous Opus, comme d’habitude. OAuth OK, abo 200 $/mois, config « propre ». 30 min → rate limit. Tokens horaires explosés. Weekly déjà -30 %. À 3 h du mat, je fixais encore le plafond en me disant : « Comment un agent peut coûter autant pour si peu ? » Réflexe d’ingé : j’ai ouvert le capot. J’ai passé le repo OpenClaw dans Code Wiki → ~200 issues GitHub dans NotebookLM. Et là, j’ai compris ce qui m’a rendu fou. Le vrai coût d’OpenClaw, c’est pas Opus. C’est le contexte fantôme. À CHAQUE message, l’agent renvoie dans le prompt : AGENTS.md, SOUL.md, MEMORY.md, skills, etc. Même pour une question triviale. Résultat : 8k–15k tokens cramés avant même que le modèle lise ta requête. Tu demandes la météo → tu payes une analyse de repo. Mais le pire, c’est le heartbeat : le cron qui « garde l’agent en vie » toutes les 10–15 min réinjecte TOUT le contexte. À chaque fois. Donc ton agent parle tout seul 24/7 avec Opus pendant que tu dors. Calcul rapide : ~12k tokens × 4/h × 24h ≈ >1M tokens/jour juste pour exister. Bienvenue au club des gens qui financent un agent insomniaque. Les 5 trucs que j’aurais aimé savoir jour 1 : 1. Opus ≠ modèle par défaut Opus c’est pour le deep reasoning / code lourd. Tout le reste → Sonnet suffit largement. 2. Layering simple = -60 à -80 % direct Aujourd’hui tout le monde flex « mon agent tourne sous Opus » alors que c’est littéralement prendre un jet pour aller au Lidl. 3. Vos fichiers perso sont trop gros AGENTS.md à 3k tokens avec des doublons, SOUL.md roman autobiographique, MEMORY.md jamais nettoyé. Chaque token ici = taxe permanente. Depuis que j’ai slimé (AGENTS <800, SOUL <500, MEMORY <2k), la facture a fondu. 4. QMD est criminellement sous-utilisé Par défaut l’agent lit des fichiers entiers pour trouver 3 lignes. QMD extrait juste les passages pertinents → ~90 % tokens search en moins. Night & day. 5. OpenClaw est amnésique par design La « mémoire » est dans le prompt, donc dépend du modèle, donc saute à la compaction. Mem0 sort la mémoire du prompt → recall auto, persistant, stable. Ça change tout. Bonus : /new est un piège Tu penses repartir clean, mais tu tues le prompt cache → tu repayes TOUT le contexte de base. /compact > /new, quasi toujours. Le vrai sujet derrière tout ça : on parle beaucoup d’agents autonomes, mais très peu du coût réel d’une intelligence locale 24/7 out-of-the-box. OpenClaw = brûleur de tokens, amnésique, accès système large. La stack qui m’a sauvé : context slim + model layering + QMD + Mem0 + approval gates. Depuis : coûts divisés par 4–5, agent stable et utile pour de vrai. Si vous utilisez OpenClaw, regardez votre conso tokens/jour. Vous allez probablement faire le même saut de tension que moi 😅

@ryanhallyall I get this for free from my local service

I know Yallcall is still making the rounds and some people in the weather space are upset about it. Yallcall happened because my audience literally asked for it. Every stream. 'Just call me when it's bad.' We partnered with a company that's done this for 20+ years to make it happen. Tons of people are already signed up and I'm proud of it. This is the last time I'll address this. I'm disappointed in some of what I've read from people in the weather community but my goal has always been to protect life and property, and that's not changing.

@CharlesChillFFB You’re paying top 5 RB prices before he even gets drafted, can’t imagine why half of this poll would be content with what would’ve been RB13. Now if that average is because he was top 3 a couple years and battled injury other years then he’ll be alright

If Jeremiyah Love gives you this average yearly line for the first 6 years of his career, are you happy with it? 204 rush 912 yds 7 TD 52 rec 401 yds 2 TD 237.3 PPR 13.95 PPG