Razzball Football Subscriptions Now Include ‘What If’ Projections

We feel we are the best in-season fantasy baseball subscription service for both season-long and DFS. It is possible that our ethos of ‘We give advice for free, avoid massive cost outlays for fancy software, and ensure our collective investment (our time, your money) goes towards creating/maintaining the best projections and powerful but simple tools’ is not for everyone. So be it.

Getting to where we are in MLB was not easy, it did not happen overnight, and it never would have happened without the feedback and support of everyone at Razzball Nation (I know that sounds like standard BS but loyal readers know Streamonator was a commenter idea – I just executed and iterated on it.).

It was about 15 months ago that Grey and I decided to invest in building the best fantasy football subscription service. We knew it would be a massive learning curve and that there are some great, established players in this space. While I imagine we will be even better 12 months from now than we are today, I highly suggest jumping on board for 2017 because things are looking really good and we need the support to justify continued investment.

Note: Free 7-day trial. You can see screenshots and more schpiel by clicking on a package

Main Group

You may select only one level from this group.

Ad-Free

You may select only one level from this group.

So what has changed from last year?

It all starts with the Data Model (aka the Pigskinonator). It is relatively easy to create weekly NFL rankings/projections with a few formulas in Excel. We have invested significant time in building a data model that strives for excellence in each component of a great NFL projection. It starts with leveraging multiple data sources to determine player roles and estimate snap counts for both the offense and defense that reflect today’s NFL. Yes, there are still a few ‘bell cow’ every-down RBs but there are also RB job shares, 1st/2nd down backs, 3rd down backs, etc. There are every-down WRs as well as WR3s who play a majority of downs and WRs who come out in 4 WR sets. Our model aims to distinguish across all those types and making sure snap counts reflect a team’s offensive philosophy (e.g., some teams are heavy 3-4 WR, some are heavy 2 TE, etc.). Player rate stats (completion rate, yards per carry, etc) are based on previous game stats (with regression) as well as player scouting grades. Team rush/pass splits are projected based on team and gameday variables (spread, home/away, etc) and carries and targets are distributed based on player role, skill, and team proclivity. Defensive player stats are derived from the same data as the offensive stats. We even built a model to identify a player’s likelihood to score TDs based on both past games, their height/weight, and Vegas points. Net-net, we strive for greatness in every aspect of the process and to be ‘smart’ enough to handle breaking changes like redistributing carries when a top running back is ruled out at Sunday 12:48PM EST.

The initial model/script evolved throughout the 2016 season as it (okay, me) got smarter about the intracacies of projecting football and understanding the strengths/weaknesses of our data feeds (aka, oh shit, nobody updates depth charts real time). It has gone through two major upgrades this offseason: 1) It can now project all 17 games versus just the current week’s game (this requires not just code changes but handling when injured players come back and some level of forecasting injuries/uncertainty), 2) Logic and data quality improvements fueled by running a season’s worth of projections, finding anomalous results, and coming up with solutions), and 3) Automating it so it can constantly incorporate the latest news as kickoff approaches.

These upgrades helped spur some major functionality ads:

• Team Pages –  I use a number of football sites as part of my research/QA and I do not think any of their team pages pack as much utility as ours. These are free until the start of the season. You can find them in the top menu. Check ’em out!
• 17 Week Player Rankings – This is where my MLB learnings/biases kick in. I HATE when other systems are too dumb/lazy/etc. to help its readers/subscribers make easy decisions. Case in point: ‘matchup’ rankings. Ooh, Week 12 is a great matchup. Real helpful – now how do I decide if great RB – bad matchup is better than mediocre RB – great matchup? Screw that. Our 17 Week player rankings show you each player’s positional ranking across every game on the calendar so you can focus on making the right roster decisions ahead of time. Better yet, these projections will be updated near-daily so injuries/role changes are reflected.
• Rest of Season Projections – Similar rationale to the 17-Week Player Rankings. Updated near-daily.
• DraftKings/FanDuel/Yahoo Lineup Optimzer powered by Daily Fantasy Cafe – Our partnership with DFC has gone well for MLB so we are extending it for NFL and NBA. Tons of functionality and you can access it on Razzball.

And that takes us to our latest creation which is ‘What If’ projections. Inspired/insisted by one of our readers, the problem this is trying to solve is ‘What is the impact if certain players do not play this week?” While the Pigskinonator is too complex (and/or I am too dense) to allow for a dynamic ‘What if’ query, we simultaneously run the projections with all ‘Questionable’ and ‘Doubtful’ players taken out of the projection system (our standard approach is to discount playing time – particularly for Doubtful). We then compare their Fantasy Points (STD, 1/2 PPR, PPR, DraftKings, FanDuel, Yahoo) against their ‘What If’ points to help identify major gainers/losers should injured players not play. How will Luck do if Hilton doesn’t play or vice versa? How will Perine fare if Kelley does not play? What happens in IDP if the team’s tackle machine is out? All of these questions are potentially solved by the ‘What If’ projections. The offense and defense ‘What if’ projections are separate reports while these projections are integrated within our DFSBot projections and Lineup Optimizer powered by Daily Fantasy Cafe (Where will it fail? Probably in a couple areas but the main one is when multiple players on a team/position are ‘Questionable’)

Here is an example of what happens to Indianpolis RB/WR/TE projections if Andrew Luck and backup RB Robert Turbin are out for Week 1.

As always, if you have any questions or ideas, use the comments or reach out to me at [email protected] or @rudygamble. Also, if you are a current MLB subscriber, you should have gotten an e-mail with a discount code for an NFL subscription. E-mail me if you did not receive it.