Solving for Points: Instructional Guide

On this page various practical approaches to using the solver will be outlined. It is worth becoming to accustomed to how you might prefer to approach the solver before playing around with advanced settings. To learn more about specific settings, check out the solver glossary

Deciding on a Solver Approach

There are three main ways to use the solver and a personal preference or specific scenario may determine which version you decide to run with.

Single-Period

This can be used to recreate the original “Transfer Suggestor”. Ideal if you just want a quick list of suggestions to consider and cycle through future GWs as necessary manually

Multi-Period

Multi-period solves are really intended to create an environment where a more coherent root move can be uncovered, through factoring in potential future moves. It can also be used to create a more literal ‘N’ GW plan

Hybrid

A mid-way between the two approaches also exists, where the first two or so GWs can be selected manually and cycled through-before closing out with a multi-period solve

Single-Period Approach

The single-period solver is the quickest to adapt to and is also how the original “Transfer Suggestor” operated. To replicate the exact same output a FT Value of 0.0 can be applied, however there is still practical advantage here in setting a value of ~1.5, to incentivise rolling transfers as needed.

Notes
  • To replicate the exact same behaviour as the old Transfer Suggestor a FT Value of 0.0 can be applied
  • Setting FT Value to ~1.5 is generally recommended for more coherent planning. (The impact of FT Value is explained in more detail lower down the page)
  • By default, the planner cycles through the GWs following the selection of a path 

https://fplreview.com/wp-content/uploads/sp.mp4

Multi-Period Approach

Deep multi-period solves are a much, much computationally heavier problems, but should generally assist in achieving more coherent plans. This solving approach considers the changes you can make in future GWs within the selected period, rather than just the selected GW alone.

Notes
  • The “Solve Depth” setting determines how many future GWs transfers will be considered in solve. You may want to hold this back from the full horizon to prevent ‘dead-end’ moves in the final GW or two.
  • Setting FT Value to ~1.5 is generally recommended for more coherent planning. (The impact of FT Value is explained in more detail lower down the page). A setting of 0.0 will generate a more literal/intricate plan.
  • You can choose your preferred line, scan the root moves for ideas, or even cycle through the transfer paths without re-solving

https://fplreview.com/wp-content/uploads/mp.mp4

Hybrid Approach

There is an approach that blends the customisability of the Single-Period approach with the power of the Multi-Period approach in reasonable time. Here we cycle through a few GWs of shallow multi-period solves, selecting our preferred moves before finally accepting a full-line to close out the plan. 

Notes
  • A solve depth of 3 or 4 GWs may be suitable here, with the first 2 or 3 GWs having manually selected paths (assuming an 8GW planning horizon)
  • Setting FT Value to ~1.5 is generally recommended for more coherent planning. (The impact of FT Value is explained in more detail lower down the page)
  • You will notice that by default when the Solve Depth is greater than 1GW, follow-up moves are automatically generated when a transfer is selected. This can be re-solved for a greater volume of suggestions.

https://fplreview.com/wp-content/uploads/hy.mp4

Understanding How FT Value Impacts Planning

One of the more confusing new additions to the site is FT Value, and the impacts and implications it brings. Firstly, it is worth understanding that there are two trains of thought regarding how we plan, and FT Value is really at the centre of this!

  1. Intricate/Literal Planning: This is where we want our plan to reflect a full set of moves over the full horizon
  2. Robust Planning: Only the live/upcoming GW is treated in a literal way- beyond this only key moves should be highlighted

A FT Value of 0.0, leaves us in the world of literal planning. A setting in the region of ~1.5, creates a baseline of what rolling a transfer and what we might expect from a typical transfer. Literal plans tend to be a lot more fragile, the approach itself has done worse in some hugely impressive analysis carried out by @FF_Trout.

This means for example, only seeing a move in one or two GWs across a solve depth of 6GWs does not imply we are burning loads of transfers. It may help to interpret it as unlisted moves occurring in a way that keeps is flexible and not tied to dependent plans.

There are times when we may want to increase/decrease that 1.5 value. For example, a few GWs before a WC, a setting of 1.0 may better reflect that there is a reduced period to find gain in.

When making plans it is really worth getting to grips with the idea that future transfers are hugely speculative and beyond certain key moves it probably best not to read into them too literally (as fun as it might be).

Activating Chips

Another core concept in planning is factoring chips. This is achieved by selecting the GW you plan to use any chip in the Team Settings panel. This can help limit the horizon used when making plans, for example we might want to play the WC chip in 3GWs, in that case we don’t want those future GWs influencing our current plan. Chips are easily factored within the planner – you can even simulate a future WC by loading a shorter planning horizon.

https://fplreview.com/wp-content/uploads/ch.mp4

Using Force Include/Exclude

Force include and exclude are great ways to put your stamp on things and let the solver build around your plans. It should be noted that for now, two main limitations apply to non-WC/FH solves.

  1. Budget Constraining booked transfers: Making a booked transfer that breaks budgetary limits is currently not handled. This will be patched in the near future
  2. Too many ‘Any’ Force Ins/Outs: In a normal GW it is suggested that only one player should be forced in/out for ‘any’ alternative. During a WC you have greater freedom with this.

Use this capability as demonstrated in the video below, please ensure forced decisions appear as a box and have been input correctly.

https://fplreview.com/wp-content/uploads/fi.mp4

Using Advanced Settings

The first thing we may want to do when using the tool is deduce the perfect settings. The default settings themselves are generated to produce a strong root move – while it is not suggested to edit these settings on a whim, there can be good reasons to change any of the settings. It is suggested to get familiar with any settings you plan to edit in the solver glossary beforehand.

 

Concluding Thoughts

All in all, whatever our approach to driving the solver, it is important to understand that the root move for the upcoming GW is where focus needs to be.  Distant moves become increasingly speculative and should be considered a such, planning in FPL needs a relatively light grip. Focus on the root moves, this might reduce our thoughts towards a pool of half a dozen options – go and research these players and the higher-level implications of these moves, such as if they create or impinge a certain key future move or factoring price changes etc. to get the most out of the tool. 

The intended use is idea generation and solve around your beliefs as a decision aid, there is no pressure or need to stick to it rigidly.

0