How to assign confidence points

Picking winners is half the game. The other half is deciding how much each pick is worth. Two players can pick the exact same winners and finish weeks apart in the standings because they weighted them differently.

Sort first, then number

Before you type any numbers, sort the week's games into three piles: the ones you'd be shocked to miss, the ones you lean on, and the coin flips. Then give the first pile 8 to 10, the middle pile 4 to 7, and the coin flips 1 to 3.

Doing it in that order keeps you honest. It's easy to hand out 10s as you go and realize later that you've rated a shaky road favorite the same as a sure thing.

Use the spread as a reality check

The point spread shown with each game is a useful guide to how lopsided a matchup really is. A team favored by 14 is a much safer pick than one favored by 2. If your confidence and the spread disagree a lot, ask yourself why before you lock it in.

That doesn't mean always agreeing with the spread. Calling an upset is part of the fun. But an upset pick usually deserves a lower number, because you're betting against the most likely outcome.

Repeats are allowed — use them

Confidence values can repeat on Pickem League. If you genuinely feel equally sure about four games, give all four a 10. You don't have to invent a ranking between games you can't tell apart.

The flip side: if everything is a 10, nothing is. Spreading your numbers out is what lets a good read on a few games carry your week.

Mistakes that cost points

A few habits show up again and again at the bottom of the standings:

• Putting high numbers on rivalry games, where records matter less and upsets are common.
• Forgetting that double-weighted games swing twice as far — a missed 10 there costs 20.
• Ignoring late news. Picks stay open until each game starts, so a quick check before kickoff can save a bad pick.
• Skipping games out of doubt. A skipped game scores zero, but it also costs you the full-slate bonus. A low-confidence pick risks very little.