Tutorial

A step-by-step walkthrough of BinkEngine's tools -- what to click, in what order, and why. This covers the mechanics of the site; for general DFS strategy (cash vs. GPP thinking, stacking, bankroll), see the How To DFS page.

Step By Step

1. Pick your Slate. On the Home page's control panel, SLATE toggles between MAIN (the full week's games, DraftKings' standard classic contests) and SHOWDOWN (a single game, Captain + 5 FLEX). Build a Main slate first if you also plan to build Showdown that week -- the Showdown game selector populates from that Main build's schedule.
2. Pick your Contest Type. GPP, CASH, SINGLE ENTRY, 3 MAX, 5 MAX for Main; SHOWDOWN GPP / SHOWDOWN CASH once a Showdown game is selected. This choice changes what the builder actually optimizes for, not just the roster count -- see "How Contest Type Changes The Build" below.
3. Set LINEUPS and WEEK. LINEUPS is how many the builder returns for GPP-style contests (Cash and Single Entry always return 1 regardless of this number). WEEK should match the slate you're building.
4. Upload your DK Salaries file. Export your salary CSV directly from DraftKings for this slate and drop it in the DK SALARIES box. This is what actually prices every player and sets the $50,000 cap -- without a real export matching this week, salaries and the player pool will be wrong.
5. (Optional) Load your own projections. If you have a paid projections/ownership subscription (LineStar-style export), drop that CSV into UPLOADED PROJECTIONS. Use the Auto / Model Only / Uploaded Projections toggle above the Player Pool table to control whether the builder uses your numbers, this project's own simulated numbers, or blends -- Auto uses your file where it matched a player and falls back to the model elsewhere.
6. (Optional) Turn on Contest Sim and Fetch Weather. CONTEST SIM models your lineup against a simulated field of a given size/entry fee to estimate real ROI%, not just raw points -- set FIELD SIZE close to the real contest's entry count for a meaningful number. WEEK SCHEDULE's "Fetch weather" pulls a live forecast that now actually adjusts passing efficiency/volume for outdoor games in real wind or precipitation, not just displays it.
7. Click Build Lineups. This runs the full pipeline -- pulls fresh data, simulates the slate, builds your lineup(s). Use Load Data instead if you just want to see projections/ownership/the player pool without building yet (faster), then Build from Current Projections once you're ready -- that skips the re-pull/re-simulate and just re-optimizes off what's already loaded, so it's fast to re-run after changing a lock, exclude, or rule.
8. Work the Player Pool. Lock (🔒) forces a player into every lineup. Exclude (✕) removes him from the pool entirely. Pool (★) is a soft preference -- the builder leans toward Pool players without guaranteeing them, so it never produces an infeasible 0-lineup build the way over-locking can. Click an FPOINTS cell to type your own projection for that player -- it rescales his whole simulated distribution and feeds the next build. "Clear Locks/Pool/Excludes/Edits" resets all of this in one click.
9. Build a Rule Set (GPP only). The rule builder under the Player Pool lets you require or limit specific position stacks (e.g. "always include QB + at least 1 pass-catcher from his team," or "no more than 1 RB from the same game"). Click "+ Create Rule" to add one. Cash and Showdown ignore this -- see below for why.
10. Check Insights before you lock in. The Insights tab has Leverage Signals (who's projected low-owned relative to their real upside -- the GPP differentiation play) and Game Environment (implied totals, pace, spread -- which games are shootouts vs. defensive slogs). Paid/Admin only.
11. Export and upload to DraftKings. The export panel gives you a CSV in DK's own bulk-upload format -- Name + ID (readable) or ID Only (DK's minimal format). Upload that file directly on DraftKings' lineup import page.
12. Track your results. After the slate plays out, log each contest you entered on the My Results page -- season, week, contest type, entries, entry fee, winnings. It builds a running week-over-week profit/loss view so you can actually see whether you're beating the games you're playing, not just guessing.

How Contest Type Changes The Build

This isn't just a label -- picking a contest type changes the optimizer's actual objective function:

• GPP builds toward ceiling and ownership leverage across multiple differentiated lineups -- ownership actively works AGAINST a player in the objective, all else equal, since the goal is separating from the field, not matching it.
• CASH / SINGLE ENTRY build a single lineup toward the highest realistic floor, and actively REWARD higher-owned players rather than fading them -- in a 50/50 or Double-Up you only need to beat about half the field, so playing the safe, popular, correct play is usually right, not something to avoid.
• 3 MAX / 5 MAX sit in between -- small, capped-entry GPP fields, so some differentiation still matters, but with far less incentive to get wild and unique across dozens of lineups the way a large-field GPP portfolio does.
• SHOWDOWN GPP / CASH apply the same Cash-vs-GPP logic to the single-game Captain + 5 FLEX format, with its own separate ownership model (Captain-slot ownership behaves very differently from FLEX ownership -- see the FAQ's ownership notes) and its own $50,000 cap covering just 6 roster spots across only 2 teams.

Free-tier accounts are capped to 1 lineup per build and Cash/Showdown Cash contest types are Paid-only -- see the FAQ for the full tier breakdown.

Player Pool Columns, Briefly

FPOINTS is this project's simulated mean projection. Value is points per $1,000 salary. Leverage flags plays where real ownership looks low relative to real upside -- the core GPP differentiation signal. Own is projected ownership (see the FAQ for exactly how that's built and its honest limitations). Ceiling (Sim P90) is a strong-outcome estimate, not a guarantee. VsPositionRank shows how a player's opponent has actually performed against his position recently -- a matchup read, not a projection by itself.

How To DFS

A general strategy primer -- the concepts that matter regardless of which site or tool you're using. If you're brand new to daily fantasy, start here; if you already know the basics, the Cash vs. GPP and roster construction sections are the parts worth actually reading closely.

The Basic Idea

You're given a salary cap (DraftKings NFL: $50,000) and asked to build a 9-man roster -- QB, RB, RB, WR, WR, WR, TE, FLEX (RB/WR/TE), DST -- that scores the most real fantasy points for that single slate. Unlike season-long fantasy, there's no draft and no waiver wire: every entry starts from the same full player pool, and the salary cap is the only thing rationing who you can afford. Whoever builds the best-scoring roster within that cap wins.

Slate Types

• Main Slate -- every early/afternoon Sunday game (typically 10-13 games), the deepest player pool and the most contest options. The default, highest-liquidity slate of the week.
• Showdown / Single-Game -- one specific game only (often Thursday night, Sunday night, or Monday night), using a Captain (1.5x points AND 1.5x salary) + 5 FLEX format. A much smaller, more concentrated player pool -- correlation and game script matter enormously since literally every roster spot comes from the same two teams.
• Early-Only / Afternoon-Only / Primetime-Only ("mini slates") -- a subset of Sunday's games, usually offered as their own separate contests. Smaller player pool than the full Main slate, but still classic multi-game roster construction, not Showdown's single-game format.
• Thursday / Monday Night -- typically single games, so these run as Showdown-format contests rather than classic ones.

The general rule: the smaller and more concentrated the slate, the more ownership and correlation dominate the decision-making, and the less raw "who projects best" matters on its own -- in a Showdown pool of ~30-40 rosterable players, everyone is looking at nearly the same short list.

Cash Games vs. GPP Tournaments -- The Single Most Important Distinction in DFS

Every other strategy decision flows from this one. Get this wrong and nothing else you do matters.

Cash games (50/50s, Double-Ups)

Roughly half the field cashes, and everyone who cashes wins close to the same flat payout. You don't need to beat the whole field -- you need to beat the median entry. That completely changes what a "good" lineup looks like:

• Prioritize floor over ceiling -- a player who reliably scores his projection is more valuable here than a boom-or-bust player with a higher ceiling but real bust risk.
• High ownership is a feature, not a problem. If a play is chalky because it's genuinely the correct, safe, obvious play, you WANT it in your cash lineup -- you're not trying to be different from the field, you're trying to beat the middle of it.
• Avoid unnecessary variance: risky punt plays, players with real injury/role uncertainty, and thin-volume touchdown-dependent guys all hurt you more than they help in cash, since one bust can single-handedly sink an otherwise-fine lineup.
• Stacking (see below) matters far less here, and a QB-heavy correlated stack can actually hurt your floor if that specific game script doesn't hit.

GPP tournaments (guaranteed prize pools, "Milly Maker" style)

Only a small fraction of a massive field cashes, and the real money is concentrated at the very top. Beating the median does nothing for you here -- you need to beat nearly everyone, which means you need genuine differentiation, not just a solid lineup.

• Prioritize ceiling over floor -- you need a real shot at a top-of-the-field score, which means rostering players who can actually go off, not just players who are unlikely to disappoint.
• Ownership leverage matters. If you and 40% of the field all roster the same chalk play, that player scoring big doesn't separate you from anyone -- it just moves the whole field up together. A correct, low-owned play that hits is worth far more to your finish than an equally-good high-owned play that hits.
• Correlation and stacking are a real edge here (see below) -- a QB with 2-3 of his own pass-catchers moves together on a big scoring day, compounding your ceiling instead of just adding it up player by player.
• It's fine, and often correct, to accept a lower floor for a higher ceiling. A lineup that busts most weeks but occasionally posts a slate-winning score is doing its job in a large-field GPP; that same lineup would be a bad idea in cash.

Roster Construction Concepts

Stacking

Pairing a QB with one or more of his own pass-catchers (and sometimes a "bring-back" -- a player from the opposing team too, betting on a shootout). When your QB throws a touchdown, that same play is also usually a touchdown for one of your other rostered players -- their good games are correlated, not independent, which is exactly what you want in a GPP where you need a genuinely big combined score, not just several decent ones.

Game Environment

Look at the Vegas total and spread for a game before assuming a player's projection tells the whole story. A high implied total (a projected shootout) lifts everyone in that game; a very lopsided spread often means the trailing team abandons the run and throws more than usual (garbage-time volume) while the favored team leans on the run late to protect a lead. This is exactly what BinkEngine's Game Environment panel and the underlying Markov simulation are built to capture (see the FAQ's "What Is A Markov-Style Simulator?" section).

Ownership Leverage

Not "always be contrarian" -- being low-owned for no reason just means you're likely playing a worse player. Real leverage is when a player's true talent/opportunity is being underrated by the field relative to how good the play actually is. That gap -- good player, public isn't fully on him yet -- is the actual edge, not low ownership by itself.

Why You Shouldn't Roster a DST Against Your Own QB

A defense/special teams unit scores fantasy points largely from sacks, interceptions, and holding the opponent to a low score. If you roster your own QB AND the DST that's playing against him, you've built in a direct conflict: your QB having a good game (avoiding sacks and picks, and his offense scoring a lot) is BAD for that DST's fantasy score, and your QB having a bad game is GOOD for it. You end up rooting against your own roster in real time, and structurally, those two players' outcomes are negatively correlated -- one doing well tends to mean the other is doing worse, which caps your lineup's realistic ceiling instead of compounding it the way a real stack does.

Bankroll Management

• Think in units, not dollars -- decide what 1 unit is (commonly 1-5% of your total DFS bankroll) and size entries off that, not off how good you feel about a slate.
• Don't chase losses with bigger buy-ins than usual -- variance is real and even a correct process loses plenty of weeks, especially in GPPs.
• Split your bankroll deliberately between cash and GPP based on your own goals -- cash games are lower-variance income, GPPs are higher-variance upside. Going all-in on one or the other is a choice, make it on purpose.
• Track everything. You cannot tell if you're actually beating your games from memory alone -- log every entry (contest type, entry fee, winnings) so your real week-over-week results, not your best-remembered weeks, tell you the truth. That's exactly what the My Results page on this site is for.

FAQ & About

About The Site

BinkEngine's projections start from a real, play-by-play Markov simulation of every game on the slate -- not a single static point total, but thousands of simulated possessions per game, anchored to that week's actual Vegas-implied team totals and pace. From there, every player's role is built from real, recency-weighted usage data (targets, carries, red-zone touches), adjusted for the specific defense he's facing and this week's game environment -- weather, injury news, and who's actually active. The result isn't one number for each player; it's a full range of outcomes -- floor, median, ceiling -- which is what powers everything downstream: the lineup optimizer, the ownership projections, and the game-environment reads. Ownership itself is modeled from how big a player's own raw projection is relative to others at his position -- the same thing that actually drives real public rostering -- then calibrated against real, logged contest results as that data accumulates, not just a guess at what the field will do. (A separate "Optimal %" signal -- how often a player is the exact salary-cap-optimal pick across simulated outcomes -- is tracked too, but that measures whether a play is sharp, not how many people will actually roster him, so it's kept apart from Ownership rather than driving it.)

Why Was BinkEngine Made

I built BinkEngine because no single site had everything I actually wanted, at a price that made sense for a guy who plays DFS seriously but not as a business. Some sites had solid projections but no real lineup builder. Some had a builder but no context -- you'd get a lineup with no idea why it was any good, what game environment you were leaning into, or whether the ownership numbers behind it were trustworthy. I wanted to see everything -- projections, ownership, matchups, game totals, weather -- laid out cleanly in one place, so that when I'm actually locking in my entries, I'm confident I'm working from the full picture, not five different tabs that don't talk to each other.

Why The Name BinkEngine

In the DFS community, when you take down a tournament, we call it "binking." BinkEngine is built to be exactly that -- the engine that gets you there.

What Is A Markov-Style Simulator?

Picture a football game happening one play at a time, computer-style. Every play starts from a "situation" -- what down it is, how many yards to go, where the ball is on the field, what the score is, how much time is left. A Markov-style simulator looks at that exact situation and asks, "based on real NFL history, what tends to happen next from here?" -- then picks an outcome (a run, a pass, a big gain, an incompletion, a touchdown) using those real odds. That outcome creates a brand new situation, and the simulator asks the same question again, play after play, until the drive ends -- then the next drive starts, and so on until the whole game is simulated. This step-by-step, "what happens next depends only on where we are right now" approach is called a Markov chain, which is where the name comes from.

Why it works: a football game is naturally a chain of situations like this -- a 3rd-and-long call is different from a 1st-and-10 call, a team trailing by 21 in the 4th quarter plays completely differently than a team nursing a lead. By simulating the game this way instead of just guessing a final stat line, realistic game flow falls out of the model automatically. If a team gets into a shootout, its QB, WRs, and pass-catching RB all naturally score well together in that simulation, because they were all part of the same simulated scoring drives. If a team falls behind big, the model naturally throws the ball more and leans on its receivers, the same way a real team would.

How is that different from a "Monte Carlo" simulator? Most DFS tools that talk about "Monte Carlo" simulation are doing something a little different under the hood: they estimate each player's final stat line on its own (say, from a statistical distribution centered on his projection), run that thousands of times, and then bolt on some extra math afterward to fake the fact that a QB and his WR1 tend to score together. Our approach builds that correlation in for free, because it never skips straight to a final stat line -- it plays out the actual game, drive by drive, and the correlation between teammates comes naturally from them being on the same simulated drives together. (We still run this whole process thousands of times per game to get a full range of outcomes, which is the "Monte Carlo" part -- Markov chain just describes HOW each individual simulated game gets built, not whether we run it many times.)

How this helps in DFS: instead of one static "projected points" number per player, this gives a real floor/median/ceiling range built from actually-plausible games, not just a number pulled out of a hat. It's also what makes same-game stacking (rostering a QB with his pass-catchers) a genuinely informed decision here rather than a guess -- the simulator already knows how often those players' big games actually happen together, because it built the games that way. That same game-by-game data is also what powers this site's ownership, leverage, and game-environment reads.

About Me

I'm Tony. Born and raised in Buffalo, NY. I'm a lifelong sports fan and, more importantly, a lifelong Bills fan -- someone who genuinely loves watching football and building lineups, not just running numbers. I've played DFS steadily since it was legalized here in New York, and BinkEngine is the tool I always wished I had.

Thank you for visiting and using the tools here -- I mean that. If you run into any concerns, have feedback, or there's anything you'd like to see added, reach out any time at BinkAdmin@Binkengine.com.

Good luck out there, and here's hoping you bink a big one. As always -- Go Bills!

Terms of Service

Last updated September 14, 2026 · Governing jurisdiction: State of New York, USA

1. Acceptance of Terms

By accessing or using BinkEngine (the "Site"), you agree to be bound by these Terms of Service ("Terms"). If you do not agree to these Terms, do not use the Site.

2. Description of Service

BinkEngine is an informational and entertainment tool that provides statistical projections, simulations, ownership estimates, and lineup-construction assistance for daily fantasy sports ("DFS") contests operated by third-party platforms (including, without limitation, DraftKings). BinkEngine does not operate, host, or administer any fantasy sports contest, does not accept wagers or entry fees, and is not itself a DFS operator, bookmaker, or gambling service.

3. No Affiliation

BinkEngine is not affiliated with, endorsed by, or sponsored by DraftKings, the National Football League, any NFL team, or any other third-party platform, league, or organization referenced on the Site. All third-party trademarks are the property of their respective owners.

4. Eligibility

You must be at least 18 years of age (or the age of majority in your jurisdiction, if higher) to use this Site. You are solely responsible for determining whether participation in DFS contests is lawful in your jurisdiction and for complying with all applicable local, state, and federal law, including registration or eligibility requirements imposed by the platform on which you enter contests.

5. No Guarantee of Accuracy or Results

Projections, simulations, ownership estimates, and all other data provided by BinkEngine are statistical estimates generated from historical and current data. They are provided for informational purposes only and do not guarantee any particular outcome, including profit, in any fantasy sports contest. Past performance of the Site's projections or lineups is not indicative of future results. You use all information on this Site at your own risk and discretion.

6. Your Responsibility

You are solely responsible for your own contest entries, fantasy sports platform account(s), and any funds wagered or entered on any third-party platform. BinkEngine does not have access to, and is not responsible for, your funds, entries, or account standing on any third-party site.

7. Intellectual Property

All content, design, software, and underlying models on this Site are the property of BinkEngine unless otherwise noted. You may not copy, redistribute, or commercially exploit any part of the Site without prior written permission.

8. Limitation of Liability

To the fullest extent permitted by law, BinkEngine and its owner/operator shall not be liable for any direct, indirect, incidental, or consequential damages, including lost profits or lost entries, arising from your use of, or inability to use, the Site or any information provided on it.

9. Changes to These Terms

These Terms may be updated from time to time. Continued use of the Site after changes are posted constitutes acceptance of the revised Terms.

10. Governing Law

These Terms are governed by the laws of the State of New York, without regard to its conflict-of-law principles, and any dispute arising from these Terms or your use of the Site shall be subject to the exclusive jurisdiction of the state and federal courts located in New York.

11. Contact

Questions about these Terms can be directed through the contact information provided elsewhere on this Site.