Weekly Feedback KPI Averages clarification?

Hi,

This thing has been bugging me for way too much time…

I’m trying to understand my latest Weekly Feedback.

I did read the various clarifications here and here.

I understand the Avg Revenue and the Avg Losses which is inline with my Average Profit (negative).

 

But… the 2 other metrics would indicate that I’m actually profitable…
Avg Loss Ratio under 1, and an Average Profit per Policy of 4.78. (positive).

I have a feeling it might be related to a skewed effect, but I can’t wrap my head around this.
 

Moreover, the baseline example had its Average Loss Ratio work as the Average Losses / Average Revenue.
 
So, what’s the issue here…
Why are my metrics saying completely different things?

 

Final thought:
I would recommend the Average Market Share to include 2 decimals (instead of 1). :slight_smile:

1 Like

Hi @michael_bordeleau

Thanks for pointing it out!

I’m looking into that odd behaviour and I will respond here once I figure that out.
This kind of thing happens because of the way averages are computed, the reason that works for the baseline is because the baseline participates in every single market in very very similar ways so they end up making a very similar loss and revenue.

But the other point on the two metrics being off in that way is interesting.
I will look into it further to see what is going on and provide an explanation below once I have it :+1:

In the meantime, please use the tables in the feedback for the profit per policy figures.

1 Like

Hi @alfarzan, i did not receive any feedback for the last 2 weeks, how can i access my feedback? For the 1st time i received an email. Thank you for your help!

Hi @yiannis_parizas3

Yes, if you go to the leaderboards for week 4 or 5, you will see your feedback once you click on “view” next to your submission :slight_smile:

@alfarzan thank you!

1 Like

Hi @michael_bordeleau

I finally had a chance to dig into this and see what is going on. I’m going to discuss only the profit per policy but the arguments apply also to the loss ratio.

How are average KPIs computed? :heavy_plus_sign::heavy_minus_sign::heavy_division_sign:

The average profit per policy is computed in the following way:

This computation is a representation of your “average market”, not your over all performance. Now what is happening in your case is that while in most of your markets your profit is positive (as seen by both the average profit per policy and your KDE plot), when you do lose money, you lose a lot. Hence your average profit is negative over all markets while your average profit per policy is positive.

Put in another way, you can see that when your profits are negative, your marketshare is also larger, hence your negative profit per policy values have a larger “weight” in your final average profit.

Why not a simple average? :thinking:

You might be wondering then, why don’t we just give you the cumulative average such that the average profit per policy would be weighted by how many contracts you win in each market.

The reason we don’t do this is because this discrepancy is informative. It is telling you that “most” of the time you are doing fine but when you lose money, you lose (relatively) a lot.

I hope this clears things up a bit more and thanks for bringing it to our attention! :rocket:

2 Likes

Thanks for the great explanation!

 

I agree :100: %

Piggy backing on this, and sharing my thought process for the benefit of other participants:

So, this submission ranked lucky 13 in Week 5…
A small average loss of 323$.

However, I’m in a situation where on average, I have a very, very , very small market share, probably around 0.2% (hidden by the KPI rounding).

Let’s just say, this situation is far from ideal.

If I want positive profit, I need to increase my prices.
My reading is that unfortunately there’s little to no room to increase my prices as I will lose market share which I have little to start with.

Need to go back to the drawing board.

 
:heavy_minus_sign: :heavy_minus_sign: :heavy_minus_sign: :heavy_minus_sign:

Through all this thinking, it becomes clear to me that there’s one KPI that needs to be public in the leader board. :trophy:

In the weekly leaderboard, right next to profit, I believe we should see the average market share for said profit, for each participant.

  • Realistically, insurers know each other’s market share so it makes this competition mimic real life :white_check_mark:
  • Moreover, it would definitely be insightful, and help participants know their gap in market share. :white_check_mark:

ie. It would help to know if the leader is at 20% market share or 2% market share and help anyone price accordingly.

2 Likes

Ah that is an interesting thought :thinking:
We’ll see what we can do about that one :slight_smile:

EDIT: And as you say you probably do need to go back to the drawing board since there is no room to increase prices!

What this result (v.low market share but positive profits most of the time) says as you highlight, is that your prices are very much in line with the market (low market share) but a little bit above it. If you increase your prices you might no longer participate, while if you reduce your prices you may go into the red further.

1 Like

I have similar struggle as yours… Low market share --> Low profit / loss yielding moderate leaderboard position. Not sure if I should lower the price more…

But having (extremely) low market share also means that the profit/loss is unstable in some sense (cannot really represent you are pricing accurately or not as it is too noisy). I think I will just try to grab more market share first, to get more sense of my expected claim bias. Maybe that’s also the rationale of most teams, which explains why almost every team is losing money.

Just to update this thread, we’ve now increased the leaderboard sizes to reduce the chance that you end up with extremely low market shares like this. Have a look at the announcement!

Hey @michael_bordeleau (salut! :slight_smile: ) ,how did you get the baseline example feedback?

That one is right here @simon_coulombe

1 Like