How Much Time Do You Need?

I promise this isn’t a bad beat post.

I’ve been betting more to reach the next DraftKings rewards tier. Obviously, this is exactly what they want: get me to wager more so I can lose more money.

More bets mean more outcomes, and lately I’ve had a few head-scratchers—bets that felt 99% certain to hit before falling apart. The funny thing about a perceived 99% chance is that you’ll never write a blog post when it wins.

With that in mind, and AI helping dig up the stats and question my percentages, here are a few that inspired this definitely-not-a-bad-beat post.

Bet 1 – Lebron James Misses by a Point

Per AI: “The parlay lost because the LeBron leg failed. That leg is unusual because LeBron doesn’t play for either team shown.”

Off to a great start.

The Setup: Maxey cleared 23 points with 1:13 left in the third. James had reached 11 with 5:36 remaining in that quarter. Philadelphia trailed 91–82 heading into the fourth, leaving me needing 44 combined points and one more point from James.

AI Probability: “I’d estimate roughly 70–80%—about a 3-in-4 chance.” A rough estimate, not historical live odds.

My Thoughts: In a competitive game, needing 44 points with James presumably playing at least eight more minutes, this felt like a winner. I would’ve put my chances closer to 90%.

What Happened: The Celtics opened the fourth with a 14–0 run and put the game out of reach within five minutes. James missed two threes before checking out with 5:35 remaining. The over hit. He never scored that last point.

Bet 2 – Williams / Burden Dominates 1st Half

The Setup: Williams had 193 passing yards at halftime. Burden had 81 receiving yards and had already scored a touchdown. I needed another 97 from Williams and 19 from Burden.

AI Probability: “My rough estimate is 55–65% at halftime—about a 3-in-5 chance.”

My Thoughts: A 55% chance when I need fewer than 100 passing yards from a guy who just threw for 193? I figured another 19 from Burden would come along if Williams reached 290. Chicago led 17–6, so more running was a concern, but this is a coach who wants to score 100. I would’ve put this closer to 80%.

What Happened: Burden suffered a concussion with 3:48 left in the fourth, finishing with 97 receiving yards. Williams finished with 279. Both fell just short. After that first half, asking for another 97 from Williams and 19 from Burden felt considerably better than 55–65%. Clearly, the past doesn’t determine the future.

Bet 3 – Burrow / Chase have Monster First Half

The Setup: Burrow had 239 passing yards at halftime, with 142 going to Chase, including a touchdown. Both Chase legs had hit. All I needed was another 51 passing yards from Burrow.

AI Probability: “I’d roughly estimate 85–90% at halftime.”

My Thoughts: How does Joe Burrow not throw for another 51 yards after putting up 239 in the first half? At 24–10, the game wasn’t a total wipeout, and he’s the current MVP frontrunner. I would’ve put this at 95%.

What Happened: Burrow threw for 32 yards in the entire second half. Thirty-two.

Bet 4 – Lawrence and Washington Connect in 1st half

The Setup: Lawrence had 197 passing yards at halftime, and Washington had 61 receiving yards, including a 50-yard touchdown late in the half. Jacksonville led 17–3 against a backup quarterback and looked like the superior team.

AI Probability: “Roughly 50–65% at halftime, with Washington’s yardage the deciding leg.”

My Thoughts: Another 29 yards from Washington would also bring Lawrence closer to the 43 he needed. The 17–3 lead meant less incentive to throw, but I still would’ve put this at about 80%. The rain was a factor all game.

What Happened: In torrential rain, Lawrence attempted just 11 passes in the second half, completing seven for 45 yards. He barely cleared his leg. Washington didn’t.

What To Make of It?

When bets don’t hit, you have to step back and ask whether you’re getting unlucky or setting unreasonable expectations.

Frankly, AI’s probability estimates surprised me. My percentages weren’t a joke – that’s how likely I honestly thought those bets were to win. Neither set of estimates came from a calibrated betting model, but the gap says something about how differently I viewed these situations.

These aren’t fringe players having unusually good games. They’re stars who need modest yardage or point totals with plenty of time left on the clock. To me, that feels like a strong position.

Evan and I joke that our bets only hit if the players have good first quarters. Apparently, now we need them to hit the entire bet in the first quarter.