https://www.chiefdelphi.com/t/mittenreview-fim-2019-week-1-preview-ready-to-launch/348346
Cycle Time
Make no question about it, this is first and foremost a cycle time game. The alliances that can out-cycle their opponents will win most matches, short of some spectacular end game shananagins or incredible defense. The goal for most teams during quals will be to be able to secure the rocket and climb RP as much as possible, so of course robots with faster cycle times will be able to fill up the rocket quicker and more consistently in matches across the length of a tournament.
Make no question about it, this is first and foremost a cycle time game. The alliances that can out-cycle their opponents will win most matches, short of some spectacular end game shananagins or incredible defense. The goal for most teams during quals will be to be able to secure the rocket and climb RP as much as possible, so of course robots with faster cycle times will be able to fill up the rocket quicker and more consistently in matches across the length of a tournament.
So let’s break this down a little, in order to achieve the rocket RP, a total of 12 game pieces need to be scored. I think that it’s safe to assume that an average alliance should hopefully be able to get 1-2 hatches scored on the rocket during auto (sandstorm), this leaves 10 or possibly 11 cycles left to do this with the 135 remaining seconds. Will a robot be able to do this themselves? Maybe, but they’ll have to be quick. The graph below gives a breakdown of the time it takes to score game pieces based on a couple sample cycle times with either 1 or 2 robots working to score them.
You can see that unless you are in the range of a 12 second cycle (get game piece, score, get game piece), filling a rocket yourself won’t be an option. With two robots working a single rocket, it becomes more manageable, allowing more than 20 seconds per cycle and still obtain the RP with enough time to get on the HAB platforms at the end of the game. I think with any week 1 and early weeks, this rocket RP will be less than common (see discussion below), but as the weeks progress then I think we will see them pretty consistently. However it also depends on how the defensive game evolves.
Average Week 1 Score
The cycle time discussion brings us to an interesting topic that always has some discussion prior to the start of competitions, which is “What will the average Week 1 score be??” Well this game is unique compared to other games, but also somewhat similar in scoring fashion to 2016. However, in this game every alliance should start off with between 6 and 9 free points to begin with since there shouldn’t be any reason to have the enormous percentage of non-working autonomous this year…let’s call it 6 and assume that 2 of 3 robots can even drive off the level 1 platform that’s it…just drive off, 6 free points. Will teams be actually scoring during week 1 during this period? Debatable. Go take a look at some early week 1 video on TBA from 2016. So let’s assume 1 game piece (hatch) on average gets scored during the sandstorm period…2 more points, we are up to a total of 8. Now, for teleop mode. In 2016, the average number of balls scored per match in week 1 was around two or three…that’s it but you were also driving over defenses. Going out on a limb here and saying we won’t see many matches with more than 5 game pieces scored during teleop, call it 3 hatches, and 2 balls. That’s another 12 points. Now we’re up to 20 total. Endgame is similar to the sandstorm period where it should be almost a guaranteed 9 points, but we know not everyone will make it back to the HAB in time, or break down. So we will say an average of 9, giving us a total of 29 points. Our prediction is that the average qualification score will be between 26-32 points in Week 1 . I hope we are wrong and average scores are much higher, but based on 2016’s week 0 to week 1 progression where the average score raised only ~20%, that lines up with the week 0 average score this year of 22 *1.2 = 26 points. Regarding the extra rocket RP, we assume very few teams will be able to complete it given that the tower in 2016 where 8 balls needed to be scored to get the RP happened in a whopping 1-3% of matches depending on event. This year, we need to score 12, split between hatches and cargo.
The cycle time discussion brings us to an interesting topic that always has some discussion prior to the start of competitions, which is “What will the average Week 1 score be??” Well this game is unique compared to other games, but also somewhat similar in scoring fashion to 2016. However, in this game every alliance should start off with between 6 and 9 free points to begin with since there shouldn’t be any reason to have the enormous percentage of non-working autonomous this year…let’s call it 6 and assume that 2 of 3 robots can even drive off the level 1 platform that’s it…just drive off, 6 free points. Will teams be actually scoring during week 1 during this period? Debatable. Go take a look at some early week 1 video on TBA from 2016. So let’s assume 1 game piece (hatch) on average gets scored during the sandstorm period…2 more points, we are up to a total of 8. Now, for teleop mode. In 2016, the average number of balls scored per match in week 1 was around two or three…that’s it but you were also driving over defenses. Going out on a limb here and saying we won’t see many matches with more than 5 game pieces scored during teleop, call it 3 hatches, and 2 balls. That’s another 12 points. Now we’re up to 20 total. Endgame is similar to the sandstorm period where it should be almost a guaranteed 9 points, but we know not everyone will make it back to the HAB in time, or break down. So we will say an average of 9, giving us a total of 29 points. Our prediction is that the average qualification score will be between 26-32 points in Week 1 . I hope we are wrong and average scores are much higher, but based on 2016’s week 0 to week 1 progression where the average score raised only ~20%, that lines up with the week 0 average score this year of 22 *1.2 = 26 points. Regarding the extra rocket RP, we assume very few teams will be able to complete it given that the tower in 2016 where 8 balls needed to be scored to get the RP happened in a whopping 1-3% of matches depending on event. This year, we need to score 12, split between hatches and cargo.
Defense
This 2019 game switched how defense is able to be played, with allowing only a single defender past the midline and with having to stay inside the frame perimeter, it makes acquiring game pieces a non-trivial task unless specifically designed to do so. With that though, the defender enjoys free reign to interfere with robots looking to score on both the rocket and cargo ship. This means that a relatively simple robot with a strong drivetrain could be enough to disrupt the scoring ability of robots. Even so, there are 12 scoring locations at level 1, and a coordinated alliance can force the defender to waste time travelling between robots. I’m excited to see how DEEP SPACE develops come eliminations begin on Saturday, and will be on the lookout for insightful and effective defense in the state this week.
This 2019 game switched how defense is able to be played, with allowing only a single defender past the midline and with having to stay inside the frame perimeter, it makes acquiring game pieces a non-trivial task unless specifically designed to do so. With that though, the defender enjoys free reign to interfere with robots looking to score on both the rocket and cargo ship. This means that a relatively simple robot with a strong drivetrain could be enough to disrupt the scoring ability of robots. Even so, there are 12 scoring locations at level 1, and a coordinated alliance can force the defender to waste time travelling between robots. I’m excited to see how DEEP SPACE develops come eliminations begin on Saturday, and will be on the lookout for insightful and effective defense in the state this week.
End Game
Ah, the end game. The part of each game it seems that can make or break the match for your alliance. A level 3 climb will nearly ensure the RP is achieved requiring only one of the two remaining robots to be on the HAB level 1. However, this is not necessarily an easy task. We look forward to seeing how many robots can successfully achieve a level 3 climb. Success in this area will be a trade-off between how long it takes to climb to each respective level vs. additional points that could be scored by cycling cargo or hatches. If your team managed to get a level 3 climb together but it takes an enormous amount of time to pull off, the alliance will have to weigh their options which are obviously match dependent. The much rumored level 3 buddy climb turns a challenging task into one that few teams in the world will be able to execute effectively. However, if pulled off, could easily turn the tide for many qual matches as well as make the difference for elims.
Ah, the end game. The part of each game it seems that can make or break the match for your alliance. A level 3 climb will nearly ensure the RP is achieved requiring only one of the two remaining robots to be on the HAB level 1. However, this is not necessarily an easy task. We look forward to seeing how many robots can successfully achieve a level 3 climb. Success in this area will be a trade-off between how long it takes to climb to each respective level vs. additional points that could be scored by cycling cargo or hatches. If your team managed to get a level 3 climb together but it takes an enormous amount of time to pull off, the alliance will have to weigh their options which are obviously match dependent. The much rumored level 3 buddy climb turns a challenging task into one that few teams in the world will be able to execute effectively. However, if pulled off, could easily turn the tide for many qual matches as well as make the difference for elims.
Quals vs. Elims
Similar to the past few years, there are no bonus points, or change in point structure associated with elimination rounds vs. qualification rounds. Simply the removal of the ranking point bonuses for certain objectives. This is important because tasks that may be more difficult but provide additional value during qualification rounds (see level 3 rocket), are suddenly less valuable during elims where an easier scoring location may be more beneficial and provide the same amount of points. We foresee scoring priority shifting from filling a rocket to try and get the RP, to filling the cargo ship. With a total of 40 scored positions (24 from rockets, 12 from side cargo ship, 4 front cargo ship) game pieces able to be scored on the field, for all but the most skilled alliances it will be near impossible to score everywhere even if you have all three robots playing offense even with the preloaded game pieces.
Similar to the past few years, there are no bonus points, or change in point structure associated with elimination rounds vs. qualification rounds. Simply the removal of the ranking point bonuses for certain objectives. This is important because tasks that may be more difficult but provide additional value during qualification rounds (see level 3 rocket), are suddenly less valuable during elims where an easier scoring location may be more beneficial and provide the same amount of points. We foresee scoring priority shifting from filling a rocket to try and get the RP, to filling the cargo ship. With a total of 40 scored positions (24 from rockets, 12 from side cargo ship, 4 front cargo ship) game pieces able to be scored on the field, for all but the most skilled alliances it will be near impossible to score everywhere even if you have all three robots playing offense even with the preloaded game pieces.
Hatch Placement
Early on in the build season there was a lot of hubub about how difficult it was to place hatches onto the cargo ship. Will we see this play out in practice? How many teams will opt to place null hatches on their cargo ship to help alleviate the potential for this issue? Many reveal videos have shown that there are many effective hatch mechanisms out there, but we need to remind ourselves that these videos are from a lot of top tier teams and may not represent the abilities of the rest of the community. Which leads us into…
Early on in the build season there was a lot of hubub about how difficult it was to place hatches onto the cargo ship. Will we see this play out in practice? How many teams will opt to place null hatches on their cargo ship to help alleviate the potential for this issue? Many reveal videos have shown that there are many effective hatch mechanisms out there, but we need to remind ourselves that these videos are from a lot of top tier teams and may not represent the abilities of the rest of the community. Which leads us into…
Ground Hatch Pickup
Much like 2017; even though all hatches are introduced to the field from the human load station, we are still seeing a lot of robots with ground pickup mechanisms. Will this year be another case where picking up from the ground will be useful early on, but taper off in advantage as the weeks go on? Only time will tell.
Much like 2017; even though all hatches are introduced to the field from the human load station, we are still seeing a lot of robots with ground pickup mechanisms. Will this year be another case where picking up from the ground will be useful early on, but taper off in advantage as the weeks go on? Only time will tell.
No comments:
Post a Comment