Ahead of season 5 starting, I’ve started testing tactics in the new match engine. I didn’t get much chance to test before the Redemption Cup and my results were a mixed bag. It was tricky to tell whether my tactics were doing well and whether any adjustments were having a positive impact over such a small sample size. Hopefully some thorough testing can give me a lot more confidence in what I’m doing.

For the first set of testing, I have played 20 training matches using the latest beta match engine, which should be the live engine for season 5. This has allowed me to complete the testing using a couple of new players I purchased before Rollover Day. The aim for the initial tests is to get a general feel for the tactics against a variety of formations. Hopefully it will identify whether the tactics need a complete overhaul and identify where to focus further testing.

My Tactics

I have ran this testing with the same 4-1-2-1-2 Wide I used in the Redemption CUp. It’s pretty light on instructions with player instructions on two players only. More positional freedom for the CAM and less freedom for the CDM. For the team instructions, we’re playing slightly wider in attack but are otherwise unchanged there.

There’s a couple of new signings in the XI; Martin Beck is a pacey striker with decent finishing. I targeted adding some real speed upfront after the Redemption Cup games and hopefully Beck will be a big improvement on last seasons forwards.

I’ve also added Timothy Cooper to play as the CAM. One thing I noticed in the Redemption Cup is the tactics generate a decent number of shooting options for the CAM. My previous options were weak finishers, but Cooper is about on par with the forwards. He has solid passing and a bit of speed as well, so I’m hoping he’ll be a well rounded option.

Results

Overall it’s definitely a positive set of results – of the 20 games played, we won 13, drew 4 and lost 3. If we were to perform like that over season 5, that would give us a very respectable 47 points. Looking at a few Ice Leagues, that point total would put us in or around the promotion play-off places which I would count as a very successful first season.

However, I wouldn’t bank on that going forward. The oppositions were all generally around our overall of 611. The highest oppositions was at 629 overall, I suspect there’s a good chance of seeing higher in league play based on some of the Spark teams people have shown on Discord. On the weaker side, there were a few teams in the 580s – I’m not sure I’ll see too many sides that weak next season.

Thoughts

For the most part the xG numbers back up the results, we won the xG battle in 11 of 20 games overall. Looking at open play xG only, that rises to 12 of 20 games. We had some really games, winning the xG battle by 1+ in 5 games. We obviously avoided getting dominated for the most part, losing the xG battle by 0.92 at most.

Looking at the performances against different formations, facing a 4-2-3-1 looks like a possible concern. The sample sizes are small for each formation faced, but the 4-2-3-1 was the only formation we failed to win the xG battle against in any of the 3 games we played. Could it be a result of the three attacking midfielders overloading our defensive midfielder? Definitely one to look closer at.

We did out perform xG from both an attacking and defensive perspective. Going forward we scored 52 from 38.05 xG, perhaps the Martin Beck effect is playing a part but that level of over performance is definitely unsustainable. Defensively, it was closer – we conceded 27 on 31.42 xG. Most likely explainable by luck/RNG.

Conclusions and Next Steps

Overall, the testing hasn’t raised any major concern with the tactics. We’ve been competitive against a variety of performances including against slightly better opposition. However, this and the other conclusions drawn must be caveated with the small sample size available so far.

The tactic is generating the majority of shooting opportunities for the two strikers and the attacking midfielder. Exactly what I want to do – generate shooting opportunities for our best finishers. The wide midfielders are averaging under a shot a game. That’s good news for Thiago Almeida and his poor shooting, we can hide his shooting and make the most of his passing skills. I’m looking to improve on White on the left side of midfield and these results suggest I can deprioritise finishing when scouting his replacement.

I want to look closer against a variety of formations before the season begins, but 4-2-3-1 will be a priority based on these results. Potentially, I could look to switch tactics to one with a second defensive midfielder. Beyond that, we didn’t see any 3 or 5 at the back formations – I’ll need to gather some data against these formations over the next week or so.

Finally, I’m expecting to see some sides significantly stronger than us in the Spark league. I want to do some testing against stronger sides and see what kind of results we can expect. I’d like to try and get a more counter-attacking strategy tested before the season.

Overall, it’s a lot of testing to get through before the season begins. Especially as I’d like to get bigger sample sizes for each test. I’m not sure I’ll get everything done in time, but I can work through things in the early part of the season as well.

If you fancy giving MFL a go. Sign up using this link and we’ll both earn a free common card when you spend $25 in the pack Store.