← CourtEdge

How to build and simulate on CourtEdge

Everything lives on one page now — the Lineup Builder — and it works the same way for every sport. Three steps, top to bottom. This page walks through each one, then covers the settings that actually change your results.

Open the Lineup Builder Open the Showdown Lab

The three steps

1
What are you playing?
Pick your slate from the chips at the top — MLB Main, MLB Turbo, NASCAR, MMA, LoL. Then pick the DK contest from the dropdown underneath. That fills in field size, entry fee, first prize and the cash line, which is what the simulation prices your lineups against.
The line under the dropdown tells you what got loaded: 9,803 entries · $18 entry · $50,000 to 1st · 1,961 paid (20%). If it says Estimated payouts, the prize ladder is modelled. See the next section to make it exact.
2
How do you want to build?
One click: Cash, Single-Entry, 20-Max or 150-Max. Each one sets the lineup count, the ranking method, the variance and the stack defaults in one go.
  • Cash — double-ups and 50/50s. Highest floor, ranked by chance to cash, no forced stacks.
  • Single-Entry — your one best lineup. Strong projection and ceiling, no unnecessary leverage.
  • 20-Max — ceiling-led, exposure-capped, stacks on.
  • 150-Max — the full portfolio. Maximum diversity, capped exposure, full stack mix.
The line underneath states exactly what will happen before you build it. If it does not say what you want, open Fine-tune.
3
Build.
One button. It builds a pool of up to 5,000 unique lineups, simulates the best of them thousands of times against a simulated field, and returns your set ranked by the method you chose. A 150-max MLB build takes about 30 seconds.

Getting exact ROI instead of an estimate

ROI is only as good as the payout ladder behind it. If you only give the builder a field size, entry fee, first prize and cash line, it models the ladder — close, but not exact, and the page will say Estimated payouts.

To make it exact, takes about ten seconds:

  1. Open the contest on DraftKings and click the Payouts tab.
  2. Select the payout table and copy it.
  3. In step 1 on the builder, open “Different contest? Enter it, or paste the exact payouts”.
  4. Paste into the box and click Use these payouts.

It reads ordinals and ranges in whatever shape DK gives them:

1st         $50,000
2nd         $15,000
4th - 5th   $5,000
26th - 200th $100

It will then tell you how many rows it read, how many paid positions that is, the total prize money, and the implied rake — and it warns you if the ladder pays more positions than there are entries, or more money than the entry fees collect.

Why bother: the difference between a modelled ladder and the real one is mostly in ranks 5–200, which is exactly where a good lineup finishes. That is the range that decides whether your ROI number means anything.

Reading the results

Above the lineups is a one-line verdict — read that first. It tells you the average expected ROI across your set, the average chance to cash, how often your best lineup wins outright, your heaviest exposure and who it is, and the stack mix that actually landed.

Then each lineup card:

ChipWhat it means
ProjSimulated average score, not the raw projection sum.
Ceil95th-percentile outcome — the score this lineup hits on a good day.
ROIExpected return on the entry fee for this contest. A * means the lineup cashed too rarely in simulation for the number to be reliable yet.
1stProbability it finishes first in the real field.
Top1%Probability of a top-1% finish. This is what large-field GPPs actually pay for.
CashProbability of finishing inside the paid positions.
OwnTotal projected ownership of the lineup. Lower means more leverage on the field.
6/6MMA only. Probability every fighter in the lineup wins his bout.
A negative ROI is not a broken number. Every DFS contest returns less than it takes in — the average entry loses the rake. If most of your lineups price negative, that is the tool telling you the slate is not offering an edge in that contest, and the honest move is a smaller entry or a different contest.

Fine-tune — the settings that matter

You do not need to open this drawer. Everything in it already has a sensible default for the build type you picked. When you do want it:

SettingWhat it does
LineupsHow many to return. Matches your contest's entry limit by default.
Rank byBest overall is the composite and the right default. Expected ROI is pure money. Chance to win for a single bullet at first place. Chance to cash for double-ups.
VarianceWidens the simulated spread. Higher leans the set toward volatile players — correct for large-field GPPs, wrong for cash.
Min / max total own %Ownership rails. See below.
Force underdogsMMA only. Minimum non-favourites per lineup.
Stack rulesMLB only. See below.

Stack rules — MLB

A stack shape written as 5-3 means five hitters from one team and three from another. DraftKings caps a team at five hitters in MLB Classic, so 5-3, 5-2, 4-4, 4-3, 4-2, 3-3 and 3-2 are the shapes that actually get built. Pitchers never count toward a stack.

  1. Open Fine-tune on an MLB slate.
  2. Type the share of your set each shape should take. Enter 30 in 5-3 on a 150-max and 45 of your lineups come back 5-3.
  3. Leave a box blank to let the optimizer choose that share freely.

20-Max and 150-Max start with a GPP mix already filled in, because that is how large-field MLB tournaments are won. Cash and Single-Entry start unstacked. No stack rules clears everything; GPP mix restores the default.

After the build, the line under the button prints the mix that actually landed — and warns you in gold if the pool could not fill a shape you asked for. That usually means the slate does not have enough cheap bats on that team to fit under the cap.

Ownership caps

Max total own % is the chalk cap, and it is the single most effective control on this page. A 150-max set with no cap drifts toward 150 versions of the field's lineup. Setting a ceiling forces the builder to find different ways to get to the same projection.

Both are applied while the pool is built, not filtered at the end, so they shape what gets simulated rather than throwing away work.

MMA — going 6/6

MMA is scored differently to everything else, because a losing fighter scores almost nothing. Six winners is not a bonus — it is close to a requirement for a top finish.

Coverage is a distribution, not a promise. On a typical twelve-fight card, 150 lineups covers roughly 90% of simulated outcomes — meaning about nine cards in ten produce at least one 6/6 lineup somewhere in your set. Zero is still possible.

NASCAR

Same three steps. NASCAR has no positions and no stacking — six drivers, one salary cap — so the stack panel and underdog control stay hidden. What matters here is place-differential: the simulation already prices starting position into each driver's outcome, so you do not need to hand-adjust for it.

Exporting to DraftKings

  1. Build your set.
  2. Click ↓ Export DK CSV next to the build button.
  3. On DraftKings, open the contest and choose Upload lineups, then select the file.

The export uses DK's own player IDs, so it uploads without re-matching names.

Common questions

Why did I get fewer lineups than I asked for?

Something you set was too tight to satisfy — usually a stack rule the slate cannot fill, an ownership cap with no lineups inside it, or too many locks and fades. The builder relaxes the minimum-difference rule before it gives up, and the build line says what it could not fill.

Why is every lineup's ROI negative?

Because that is often the truth. A contest pays out less than it takes in, and the average entry loses. Positive ROI means the tool thinks that specific lineup beats the rake in that specific contest — it should not be the normal case.

Why does the same slate give different lineups each time?

The pool is built with randomness on purpose, so you get genuine diversity rather than the same ten lineups with one player swapped. The ranking is stable; the pool underneath it is not.

What does the * after an ROI mean?

That lineup cashed in too few simulations for its ROI to be trustworthy yet. The number is shrunk toward zero rather than shown at full confidence.

Where did the PGA pages go?

Retired to the archive. Old links still work — they redirect there rather than 404.

Open the Lineup Builder See what's live today