A Grinder's Story - Episode 3
Good afternoon! I have some results, observations, and next steps to share today and a bonus feature for the blog. To avoid you skipping directly to the charts below I will give you a quick summary and then dive into the details:
- spoiler alert I was profitable in MLB for the last week
- The observations are interesting and thought provoking
- The next steps are fluid and surprising
Now for the bonus feature of the blog – Q&A. I am fairly active in the RG discord channel and post links to my blog in those to share and get feedback. I am frequently asked questions in the chats or via DM. I love the questions because it gives me direction on information I should have put within my blog and also challenges me to consider my process. Please feel free to reach out to me via the comments below or via discord. You can find me in discord as Hutchboy180. To preserve the integrity of the dialogue I am cutting a pasting directly from the actual interactions so please forgive spelling errors or shorthand.
The first question came from gregr1224
*Q – why no triple ups, quints or boosters in your game selection? especially if you’re staying off the opto train?
*A – great question…..Firstly I’m not committed to being off the optimal LU just yet I’ll decide that daily after research and review. for contest selection, I’m working through what works best. I may land on some of those games as I figure out the best mix for me. I can tell you that those contests had a very low ROI for me in NBA they may become part of my contest mix for MLB but i’m working through it and keeping the changes and contest mix as simple as possible so I have less noise in the numbers. Baseball is a long season and I will find the best mix for my play style but it may take a few weeks i became very very comfortable with my NBA mix and it showed sustainable results i’m just trying to build up to that in MLB
To expand on this answer a little bit, I’m going to speak about process improvement. I’m not going to make anyone a Six Sigma black belt or Kaizen expert (are you impressed? I actually have quite a bit of experience in my former professional life with these formal process improvement systems). A basic concept that anyone can understand revolves around a concept referred to as “noise” in the numbers. Simply stated if you are trying to improve a process, the best approach is to remove as much variance as possible so you can isolate the impact of the remaining variables. This applies to the question and my current efforts by stating that I am in the middle of assessing what contest types to enter and I am trying to limit the changes so I can determine which contests are having what type of impact.
The next question came from Lyanetsko
*Q – great read, I respect your honesty and transparency. I have a question and a comment on 2 of your Key Learnings..on your first one “I have to do more individual research on players and matchups on my own and become more consistent in my approach to building lineups.” What research do you think you’ll do that the BAT doesn’t already account for in it’s projection?
2. An incredibly high number of players in my H2Hs are all using the same optimal lineup from The Bat projection system
I noticed this alot even with my 50 1$ h2h’s can anybody attest to this being normal? like out of 50 h2h’s 20 are just dups with the bat optimal
*A – I’m not going to pretend I can out research the bat or Carty but on full slates there are dozens of very very close calls and different lineup construction decisions to make as well as ownership considerations. additionally, I will want to increase my tournament exposure at some point and spending time at a player level can help with those decisions
I guess it’s like baking a cake from scratch vs using a box mix. In theory, the box mix has been scientifically measured and quality tested and should be “optimal” but having a knowledge and and experience making from scratch can be very useful even if you are using a mix…..idk if that analogy lands but it’s the best way I can say it at the moment.
Thats also why i will be tracking daily my results and the bat optimal – it’s a long baseball season and if the data shows that I am consistently under performing against the bat Opto #1, i may make the decision to just use it every day and move on to other decisions and strategies. I’m in the experimenting phase right now.
I actually have quite a bit more to say about this topic but I will save that for my closing thoughts and although it’s not intended to be, I am sure it will be controversial. (Shameless mid article teaser to keep you reading)
Now let’s dive into the results:
Ta-Dah! All in a profit of $131.59! Okay, I get that it is a little underwhelming, but you have to remember a few things. I adjusted the process and any time you make changes there is a little learning curve. What is encouraging is the favorable trend that occurred the last few days of the week and it’s always good when the hills are higher than the valleys.
Don’t completely judge me on my commitment to reduce my daily MLB buy-ins. I should have stated Slate Buy-ins as several of these days had multiple slates with enough games to warrant playing them as stand alone featured slates. In my observations I will explain a bit more about why I was investing more than planned as well.
You can see from the chart above that out of the 7 days I had:
- 2 really good days
- 3 days that would qualify as grinder break even to me with the loss being less than 10% or less than $100
- 1 “bad” day
- 1 day I don’t even have a name for …. I will just call it stupid tax day – where you pay an arbitrary fee for your commitment to stupidity
Finally here is the data I committed to tracking. For simplicity, I am only tracking one slate per day although I may play more so these numbers aren’t meant for direct correlation to my daily profit results although they are still directional. If you are interested in the full spreadsheet feel free to reach out and I can send it to you. For the blog I’m going to limit it to the week we are reviewing.
This week we see the delta between cash and GPP have a more traditional spread with only 2 days where the gap was less than 10 points. I’m missing data for one day of the bat opto hence the black fill in although I am confident the bat opto outperformed my lineup. There is slight significance to the fact that on my 2 “good” days, my lineup outperformed the bat, but more significantly crushed the average minimum cash line.
Key Observations/Takeaways for the Week
- Focus and Attention although difficult to quantify are finite resources.
- Although most of my player level research was right in line with The Bat, it was very useful for making key decisions and building SE lineups.
- To “bat opto” or not “bat opto” is no longer an unanswered question for me (I’ll save this for the end as a stand alone commentary)
- DoubleUps and SE contest mix working well.
- Ownership creates opportunity even in cash games. (more detail here included in the final comments)
Let me unpack that first one with some context and detail.
Focus and Attention – I am still actively playing NBA DFS since MLB season started. I will not post those charts right now because even my transparency has some shame limits 😊. Let’s just say that since the beginning of MLB, my NBA performance, which had been very stable and profitable became the very opposite. Many of you who routinely play me in H2Hs have certainly noticed as I am a card-carrying member of donation season the last couple of weeks. The sharp players like dannydonahue have actively been seeking out and buying my H2Hs like they were exclusive Top Shot moments and I don’t blame anyone for it. I understand variance but my performance has not been that. My approach to contest selection was constant and the variable was in the product I was producing from lineup building. That was a direct result of trying to manage MLB and NBA and compounded by trying to do both with cash games and 150 max MME.
Fast forward to this past week and although I was not profitable on NBA, I did experience an upward trend from the rock bottom results I had been experiencing. This culminated in very profitable days on ……. Wait for it……you guessed it (I hope) – both days I won at MLB I also had profitable days in NBA.
So what changed? Well the changes I made to MLB last week eliminating 150 max helped significantly by simplifying my builds and narrowing my focus. I noticed the impact early in the week even though I wasn’t profitable yet, I felt more comfortable and less rushed. I implemented the same approach in NBA and reduced my MME exposure from 150 max to 20 max that I had been doing in NBA and am very comfortable with. The real lightbulb moment was on Monday, the 12th and Tuesday, the 13th when the NBA slate started at 7:30 – 30 minutes after MLB. I’m not sure how other people are managing both sports with a full compliment of contest types and entries, but it’s not something I can do right now with success based on what I have seen in my results.
- I am happy with the favorable trend so I don’t want to make significant changes to much and see if I can sustain and expand on this success for a week.
- On days when the NBA/MLB slates both start at 7pm, I will not play a full complement of contests on both sports. I will fully play MLB assuming it has over 4 games on the slate and limit my NBA exposure or not play NBA at all. I haven’t looked to see how often this will occur but I do not expect a conflict of this type every day. As I mentioned above, this is why some of my investment on MLB remained high as I already slowed down on NBA last week when I saw the impact.
- On days when there is not direct overlap between MLB/NBA, I am going to enter 20 max for MLB in the $4 and below contests. I will continue to enter Cash games and DUs as before. I can review the performance on these next week.
As always, I appreciate anyone who took the time to read and share in my ongoing journey. I do love the questions and interactions so again please leave a question below or reach out to me. I will ask for a “like” if you are so inclined as it has to matter to someone somewhere I guess?
Okay – feel free to let the above be the ending to this episode because the rest of this is perhaps not for everyone. I have a small circle of DFS friends I chat with on a daily basis – big shout out to my boys Luka77 and Billsbestone who started a little community with me. They have encouraged me to share my thoughts on the subject of blindly using an optimal lineup produced by some combination of projections and optimizer.
I want to get a couple of things out here first.
- I am not saying what is or isn’t right/correct
- I don’t disagree with what anyone else is doing and, in fact, I absolutely see the approach and agree with the underlying reasons and math behind the decision.
- I can’t at this point show any data or results to “prove” what I am planning to do is profitable or not. No matter the outcome, I am not going to connect results to this strategy because I’m not in the business of proving I am right or anyone else is wrong.
As the Sick Puppies say in the song anthem of my life, Riptide “I won’t justify The way I live my life ‘Cause I’m the one livin’ it”
It is not my intention to play the bat optimized lineup in cash games going forward. The reason is based on some assumptions and impacts to financial results that I will lay out below.
Here is what the reason isn’t:
- Not a criticism of the bat projection system as I will be using it as always as my primary resource for building lineups
- Not because I am confident I can produce a higher scoring lineup than the opto to any certain percent of the time
- Not a best practice I’m sharing or encouraging anyone else to do
On any given MLB slate, I am running into 25-40% of duplicated lineups in my H2Hs when I use the bat opto (assume I mean the top projected opto). While I am aware that doesn’t mean all of those are users who are blindly playing the bat opto, my duplication rate on H2Hs when I don’t use it is less than 5% and usually 0 on a slate of any size. Additionally, in SE double up contests with 500-1000 entries the lineup duplication ranges from 8-15%. I also see duplication in other contests but these are the bulk of my cash game exposure.
My lineups were so bad to start the season that nothing else really mattered. What really stood out this week was the impact on my results when my lineup was different than the opto and finished with more points. Obviously when your players score fantasy points you hope and expect your “winnings” to increase. It’s something you don’t notice quite as dramatically in the NBA for example because points are scored so quickly. In MLB, fantasy points aren’t usually accumulating with consistency other than pitchers and even then are accumulating more in batches or tied to events ie a strikeout, a stolen base…etc. Several times this week I was able to watch in real time the impact on my winnings when my lineup score passed the opto or was passed by the opto. Often it occurred when the lineup wasn’t near a contest cashing line so the impact was isolated to just that variable. I’m not going to assign and number or value to it because I didn’t track the data. Although I am systematic and data driven, some assumptions and variables involved here keep me from even trying to say this is a purely math decision with a predictable outcome.
Here is my core hypothesis. I believe that the upside to scoring more points than the opto is enough to play a lineup in cash games that is different even if it projects for less points, primarily because the variance in baseball is so significant. I don’t have an assigned number of points I’m trying to stay within so please don’t ask me. I don’t have fixed rules on what that different lineup will look like or should look like. Generally, I expect my lineups to be fairly similar to the opto but that will not always be the case. I am going to intentionally look to change the player(s) I feel have the highest variance or have a pivot(s) with very similar projections. Again, I don’t have specific rules but I will certainly discuss some of these decisions in later blogs.
“Beating the Opto” isn’t the goal but not playing the opto creates a floor in cash games that was very useful for me in NBA DFS. I play H2Hs largely so that on the days I have a bad lineup I still have the opportunity to return some of my investment by having a less bad lineup than many players in my H2H contests. I get that I could just play a higher volume of H2Hs and ignore the duplication but to do that I would have to play higher $$ H2Hs and generally, the higher the buy-in, the better the aggregate skill level. Plus, I have the advantage of knowing the exact lineup of 25-40% of my opponents. There is still no guarantee of success but I would like to think it’s an advantage in DFS knowing what the opponent’s lineup is. In some cases, I may even have the flexibility to swap players during the slate in a way that guarantees a win if the opponent doesn’t swap.
And Finally, on the days I score less than the opto, that doesn’t mean I can’t still do very very well by cashing in DUs and other contests I have entered with my cash line up.
I’m sure that won’t be the final word on the subject but as Forrest says, “That’s all I have to say about that”. Again my thanks and appreciation for being a part of my story as the Grind goes on….