Morgan Stanley's Overwatch League Research Is A Joke

MORGAN STANLEY'S OVERWATCH LEAGUE RESEARCH IS A JOKE

That's not anyone from Morgan Stanley. It's from "The Big Short" (Photo: Plan B Entertainment)

That's not anyone from Morgan Stanley. It's from "The Big Short" (Photo: Plan B Entertainment)

TNL Take:  Like everything in life, people always want to know the future: Is she going to like my Instagram pic? Will my son get into Princeton? Are the Warriors going to blow it again this year? Can I avoid any future meetings near Times Square?

The same thing goes for eSports.

For something that’s really just been born, people already want to know what it’s going to be when it grows up.

The reality is that it’s really, really, really hard to predict.

Yet many people are in the business of doing just that. I’ve made it clear how I feel about market estimates many times but a recent study by Morgan Stanley – although it’s still early in 2017 – could win the WTF Of The Year Award™.

[Note:  I reached out to the Morgan Stanley team that put together the report for comment on how they came to their estimates.  I was passed to their compliance team who said they would not comment on the data in the report or for this article]

Morgan Stanley’s report released at the end of March was titled:

Morgan Stanley eSports Research Report (Photo: Morgan Stanley)

Morgan Stanley eSports Research Report (Photo: Morgan Stanley)

But really focuses on the revenue potential of Activision-Blizzard's upcoming Overwatch League, aka the NFL of eSports.

 

Let's go through the divine comedy that is the Morgan Stanley study.

/01 REVENUE VARIANCE

The first thing that stands out in Morgan Stanley's report is that their Revenue estimates range from their “Bear” annual scenario of $20M to it’s “Bull” scenario of $720M.

Let me repeat that.

Between their worst and best case scenario is a difference of $700M.  If I did an analysis and gave a similar variance of $700M to my manager -  my work email would more than likely not be working by the end of the day.

Even the base estimate or "take your best guess" number is $100M - which is 7X lower than the best case scenario.  

That's a lot of spaghetti against the wall.

 

/02 AUDIENCE VIEWERSHIP

Morgan Stanley's base case estimate also assumes that each regular season game of the Overwatch League will average 72,000 viewers and 7.7M viewers for the playoff rounds.

How did their magic 8-ball come up with this number?

By taking a 19M MAU (Monthly Active User) figure - which is very high considering that Overwatch just hit 30M registered players - and taking "40% of those people watch eSports" and coming to the 7.7M figure for the playoffs.

If Riot's League of Legends massive World Championship Finals 1 day event really only draws ~7M US viewers, how is a game that's only about a year old going to draw more viewers across multiple playoffs and Finals games? 

TNL Infographic 028: 2017 Finals: Sports vs. League Of Legends Viewership (Infographic: The Next Level)

TNL Infographic 028: 2017 Finals: Sports vs. League Of Legends Viewership (Infographic: The Next Level)

So what about the 72,000 per regular season game?

By then taking the "magic metric" of 0.9% of the 7.7M playoff viewers to get to the 72,000 viewers across three digital channels - and considering how close Blizzard is with Facebook and their recent Exclusive deal, 3 digital deals seems high as well.

 I have zero idea where the 0.9% metric comes from but I do know data.

Here are the top Overwatch events by Average Viewers on Twitch since the game launched:

Overwatch Atlantic Showdown (Photo: Waypoint Twitch Dashboard)

Overwatch Atlantic Showdown (Photo: Waypoint Twitch Dashboard)

Overwatch Open (Photo: Waypoint Twitch Dashboard)

Overwatch Open (Photo: Waypoint Twitch Dashboard)

Not one major event on 1 digital platform has hit close to the 72K average viewer estimate. Getting to that total maybe potentially possible if 3 digital platforms deals are struck but it's still not accounting for two factors: Number of regular season games and International Audience %. 

The first is that not only does that 72K average need to be achieved but sustained across a much longer season than these shorter major events. The second affects the next the point, Ad Revenue.

 

/03 AD REVENUE

Morgan Stanley’s base case estimate also assumes that the majority of revenue – 62% will come from Advertising/Distribution (32%) and Sponsorships (30%). That means that 62% of the base case $100M revenue estimate will have to come from Brands.

That’s $60M+. Per Year. For One League. For One Game.

Further, the math used to come up with the $32M doesn't factor in the International viewing audience - I'd guess at least 1/3 - and a high average CPM, or the cost Brands pay per video ad.

The $30M+ expected for Sponsors like a Coca-Cola, McDonalds, Ford, etc. is potentially in reach if there are 16 teams by this Fall and the viewer numbers beat the current estimates.

Here's how Overwatch performed on TV last year:

TNL Infographic 020: 2016 eSports TV Ratings (Infographic: The Next Level)

TNL Infographic 020: 2016 eSports TV Ratings (Infographic: The Next Level)

 

For those keeping score: that's high audience and high ad expectations.

 

/04 TEAM PRICE

The report also estimates that Overwatch League franchise prices would be ~$3M each.

No way the price will be this low.

I spoke further about all these points on SiriusXM's eSports show the day the report was released:

In a somewhat surprising move, a few weeks after Morgan Stanley released the following "update" a few weeks later:

Morgan Stanley Update To Initial Overwatch League Report (Photo: Morgan Stanley)

Morgan Stanley Update To Initial Overwatch League Report (Photo: Morgan Stanley)

Here's what Morgan Stanley learned after apparently actually speaking to people in the industry:

/01 The $3M per team cost they "assumed" seemed low based on the wide variance of prices rumored, which is on the much higher side.

/02 Traditional sports teams are the targets to buy slots in the league.  Based on the little that's been released publicly, that's more than obvious.

/03 I'm just going to quote the headline they said for this point and translate:

"Non-Endemic Advertising/Sponsorships Will Require Customization and Investment from Brands/Agencies" or in other words: "Whoops, Our Ad Revenue Estimates Were Probably A Bit High. Sorry About That".

/04 Overwatch needs to become a better spectator sport. Or, their viewership estimates maybe inflated.

/05 Amazon Game Studios with Breakaway and Twitch are doing some cool things. 

 

But they didn't address the key data points and how they got to their original estimates.

 

In the Gaming world, games like Rocket League, H1Z1 and this years PlayerUnknown's Battlegrounds are titles that weren't marketed as eSports titles but are now very popular.

Here's April's top Twitch games by Hours Viewed:

April Top Twitch Games By Hours Viewed (Photo: Gamoloco)

April Top Twitch Games By Hours Viewed (Photo: Gamoloco)

I'm rooting for the Overwatch League to bring some much needed structure, support and a large company running the show - I just hope that the early expectations are tempered.