Sunday, 3 January 2021

Gravity (NASDAQ: GRVY) - Ragnarok Online, from PC to Mobile

Gravity (NASDAQ: GRVY) is a South Korean video game developer, mainly known for Ragnarok Online (RO). RO was launched in 2002 and was pay to play, but had their code leaked/hacked leading to the birth of private servers.
What started primarily as a PC game, majority (~70%) of their revenue now comes from Mobile games. GRVY's increase in revenue in 2018 came primarily from the launch of their Mobile games, nearly tripling and doubling their revenue in 2017 and 2018 respectively.

The stock price has huge run up in 2020 thanks to their impressive growth in their mobile games and improved margins.


Ragnarok Online (PC)

Gravity runs their own servers while also licensing out to various publishers. Most servers are free to play / freemium while allowing players to use real money to purchase items or a VIP subscription to speed up their progression. Gravity also made some questionable decisions such as shutting down the European server instead of coplying with the GDPR. For some reason, RO is extremely popular in Thailand. 

Revenue from this segment has not been growing much, maybe simply due to the game's age or lack of appeal to modern audiences. Gravity made some changes and launch some new "classic" servers as well but the pay to win features still remain. Some turn to private servers for the classic experience or because there's no P2W (though some servers do have P2W in them).


Mobile has been a cash cow for Gravity and the primary driver for growth over the past few years.

Some features that keep players playing.

  • Daily/Weekly tasks to keep players logging it "just for the dailies"
  • Auto Battle for those who feel can't keep up because of real life commitments

Pay to Win/Convenience

  • Enchanced Rewards for Daily Tasks (Advanced Kafra Log), ~7 - 24 USD/month
  • Monthly Premium "Subscription" that can also be sold to other players for in-game curency, ~10 USD/month
  • Gacha system for equipment, required for PVP (Player Versus Player) or War of Emperium (WOE / Guild vs Guild)
This is not extensive and there are probably other features they use to monetize the game. Most (mobile) games rely on whales for a substantial amount of their revenue.

On the other hand, loot boxes in video games have been getting increased scrunity from governments over their gambling-like nature.

Licensing/Royalty Income
Gravity makes money by licensing out their games out to other publishers, or allowing other games to include items from their IP.

Gravity is launching several mobile games in the 1H of 2020, with Ragnarok X and Action RO2 being mobile MMORPGs.Will they be able to repeat their sucess with their previous launches? Or will a new MMORPG simply take away the existing player base from their existing games?

Valuation wise, GRVY trades at a P/S at 3.99. Zynga is probably the closest comparison, being a mobile-only developer/publisher which trades at a P/S of 6.05 and is not profitable.

YoY growth looks a lot less impressive than other video game companies. A lot of it seems to be baked in expectations of the sucess of their mobile games given the bad first quarter of 2020.

Yahoo Finance

Rather than looking like a "monster run up" on the chart, it looks more like a regression to the mean. Q3 was definitely impressive but a question is whether it can be sustained. The price run up after Q3 looks like its pricing in another killer quarter. 

The RO IP isn't that popular outside of people who played the original games and the MMORPG genre is less acessible to people unlike Candy Crush or Puzzle & Dragons, so drawing in new p(l)ayers might be harder. The Ragnarok IP is more like World of Warcraft (WoW), popular with those born in the 80s or 90s but doesn't attract much attention otherwise. While I suspect that Q4 will be a good quarter, I question how much more revenue can continue growing and whether the P2W model will continue to be sucessful. Most MMORPGs see declining player base over time, even with updates. 

Is Gravity's solution of launching a new game and catcing whales every few years going to be successful? Or will they alienate their playerbase and split the community between their old and new games?

Saturday, 24 October 2020

Limelight Networks (NASDAQ: LLNW) drops by 30% after Q3 2020 results


Source: Finviz

Fastly (FSLY) also dropped by 30% after their earnings revision.

Source: Finviz

Comparing the P/S before their respective drops, LLNW was trading at a much more reasonable valuation.

Data from RocketFincnail
For a stock trading at a P/S of 3, a 30% drop for missing EPS estimates by $0.03 seems to be an overreaction.

EPS alone was hit by $0.02 because of the convertible note offering that was made in July.

Earnings Call (Source:

Operating expenses increased approximately $600,000, primarily due to one-time costs related to organizational realignments

Revenue in the third quarter was $59.2 million, up 15% year over year.It marked our highest third quarter ever, and second highest in company history.
EBITDA was $3.7 million, up a very strong 49% over the third quarter of last year. Adjusted EBITDA was $5.6 million this year.

EBITDA may have been fine to use before given that that their equipment and hardware have depreciation. However, they now have a huge interest expensive (no) thanks to the convertible notes and is no longer a good measure of performance. Interest on the convertible notes alone cost them about 962,500 a quarter (110 million * 3.5%).

We are leaving our full-year guidance unchanged except for GAAP and non-GAAP EPS, solely due to the impact of our convertible debt offering, and the additional cash and non-cash interest expense. Interest expenses will have a $0.03 impact on GAAP EPS adjusting to a $0.03 to $0.13 loss, and non-GAAP EPS is adjusted to a $0.02 loss to an $0.08 gain. Annual cash interest expense from the July convertible debt offering will be $4.4 million, payable in February, and August. Non-cash interest expense for the fourth quarter of 2020 will be approximately $1 million.

No change in FY guidance.

In the July call, the customer that we were hopeful that we were on the one-yard line of closing it. Actually hasn't closed, and may not close 'till the middle of next year, due to some geopolitical things that are going with that. I've gone on with this company.

Hints that ByteDance/TikTok is their potential 'outsized' customer.  

And so with that convertible market operating efficiently as it was, we felt that it was appropriate time to take advantage of that market, and then do something with that cash. And whether that's just going to be internal improvements, added capacity, accelerated capacity, moving into adjacencies internally with added R&D expenses, and sales expenses, I think it's going to be a combination of all that. And then it gives us the wherewithal if something comes around that we want to go out and pick up on the M&A basis that we have that flexibility in order to do that.

Nothing planned with the convertible note offering, this is very disappointing especially when it hits their bottom line a lot.

Unfortunately, I'm still holding shares and have some 20 Nov puts that are probably going to get assigned to me. I still have 20+ days and the $5 puts shouldn't be that bad after assignment. The $6 puts are going to hurt though. Hopefully I can close them if IV drops. FSLY reporting worse than expected results could be be positive for them since they did much better despite demand dropping, or may end up dragging it down it together.

While the results are rather disappointing, I don't think it warrants the 30% drop given the current price and valuation. This is a sector with a lot of tailwinds but the management hasn't managed to meet their guidance. Its also kind of scummy that they released the "Streaming Video Viewing Reaches All-Time High According to Limelight Networks Research" press release a week before the results were released.


The growth in ex-US markets also present opportunities for growth. With streaming wars taking off, LLNW presents an opportunity to take part in it without having to choose a single company.

Back in 2006, Akamai (NASDAQ: AKAM) wanted to acquire LLNW and Tuition Build in 2014 for 645 million. The current market cap at 4.20 is about 512 million which is even lower than that. Maybe Amazon, Microsoft or Google can acquire them since its just chump change to for them at 2-3x of the current valuation.

Sunday, 4 October 2020

BlackBerry (NYSE: BB / TSE: BB) - Worthy SaaS play or forget it?


  • BlackBerry is now a Software as a Service (SaaS) and Licensing as a Service (LaaS) focusing on Cybersecurity and automotive software with >70% gross margins
  • QNX is a big player in the automotive and EV industry
  • Software and Services revenue growth is not convincing enough

BlackBerry is a shadow of its former self and the revenue decline from its peak speaks for itself.
Figures from
In recent years, they've been become more of a Software as a Service (SaaS) and Licensing as a Service (LaaS), focusing on Cybersecurity and automotive software via QNX. While the company focused on more software offerings, the legacy business via licensing still contributes a large amount of the company's revenues.
Source: Company's Presentation
Digging into their older results, they've managed to pivot away from the hardware business and grow their S&S revenue. Revenue mix has changed over the years but its not easy to decipher as the classifications have been changing.
Source: 10-Q and 40-F filings
Chart: Yahoo Finance
Ignoring their licensing revenue, Software & Services (S&S) growth is practically non-existent in the recent years. The biggest growth for (S&S) was in 2016 and the market largely ignored it, probably due to declining hardware revenue which still contributed too much of their revenues. BB announced a lot of new partnerships and clients in their press releases but they haven't seem to be reflected in to their results.
Source: The Motley Fool
BlackBerry Spark contributes majority of their S&S revenue but it is very to get an individual breakdown to tell the true growth of their individual offerings.
Q1 & Q2 results
Revenue growth for the latest quarter was fueled by licensing rather than software growth, and while S&S revenue declined.

While QNX took a hit due to the decline in automotive production, its a play on the autonomous and EV industry by providing the software and infrastructure which is integrated with their other offerings, rather than the self-driving capabilities. Maybe Tesla could open up their software offerings but I doubt it since its more of build-to-suit rather than a one-size fit all option. Some new developments include XPENG (NYSE: XPEV), Canoo (NASDAQ: HCAC (pre-merger) and the partnership with Amazon's AWS so its not just the legacy players. Amazon's investments in Rivian makes this a very interesting play as well. QNX faces competition from AGL, but an "out of the box" solution and proper safety certifications probably works better for automotive companies. Unlike Tesla, most of them aren't technology companies and definitely not on the software side.

Finally, CEO, John Chen is an advocate of co-operating with governments when it comes to privacy seen here and in his blog post in 2015. It might not be in the best interest for individuals but I can see this making them to be the preferred partner of governments. The S&S revenue growth is not convincing enough to invest in even at the P/S of ~2.7. However, if you consider their licensing revenue as a driver of growth, it might be worth looking into but I feel that there's too much legacy and uncertainty in terms of recurring revenue or growth.

Cybersecurity is rather CRWDed (get it?) area right now and there more attractive stocks out there. For people not the industry or without knowledge, its rather hard to understand the underlying businesses or technical advantages that one has over the other.

The opportunity in BlackBerry for me looks to be writing puts or calls assuming this ~4.40 is the new floor and barring any overall market decline. I don't see any bad news that would adversely it affect nor do I see news that would send the stock exploding to new highs. ~14 DTE 4.50 put at this levels give you about $10 for $440 risked for cash secured,or about 2.27% every 2 weeks.

In the meantime, I'm looking forward to the phone coming out in 2021.

Sunday, 27 September 2020

Rethinking My Portfolio & My Watchlist

I use Portfolio Performance to track my portfolio, its free and open source and has some pretty nifty charts & calculations. At the start of the year, my portfolio looked like this which basically all REITs.
In February, I sold off most of my holdings, avoiding the crash for Suntec though FLCT and Ascendas have already recovered to their previous highs. I didn't really take much advantage of the crash and started buying UOB too early and am currently sitting on a ~6% loss even after the dividend payouts. I managed to buy DBS around $18 and already sold it for a nice ~20% gain.

My current portfolio looks like this, with the cash portion that is also being utilized for my cash secured puts.
Shifting away from Singapore, if we look at ACWI; which tracks the large and mid-cap of developed and emerging markets using the free-float market cap weighting, Singapore is only 0.3% of the fund. It doesn't really make much sense to be overweight Singapore just because of my home country bias or foreign exchange risk. SGD is also just a basket of currencies anyway, so just invest in a basket of stocks that are denominated in basket of currencies and you will have reduced exposure to currency risk.
ACWI Prospectus, page 21
I think SGX is only compelling if you are willing to invest in the "tech" or proxies to them such as iFast, Venture, UMS, Frencken, Micro-Mechanics or AEM. Otherwise, there's not really much growth and really more on dividends and most of them are recovery plays right now. While I'm only left with UOB, I'm probably going to divest it for a combination of XSOENTSX and VXUS so that 50% of my holdings in global exposure.



NTSX is the WisdomTree 90/60 U.S. Balanced Fund which uses treasury futures to get a 1.5x leverage on a traditional 60/40 ETF, making it a 90/60 portfolio. It has historically outperformed the SPY and the 60/40 portfolio as well. There's a very long thread on thread on using a mix of leveraged index and bond ETFs on the Bogleheads forum about this and a backtest on portfolio visualizer found in the thread.

XSOE is the WisdomTree Emerging Markets ex-State-Owned Enterprises Fund. It's basically an emerging markets ETF without government ownership (>20%). There's also CXSE which tracks Chinese companies excluding state-owned entities. Historically, these ETFs outperformed those with state owned entities. Given the current situation with government bail outs, I feel that they are more than likely to continue to outperform since state owned entities will be forced to do 'national service' (think SIA or Sembcorp Marine).

VXUS is the Vanguard Total International Stock ETF which tracks the total world excluding US. Its only here so that I have exposure to Europe and Japan for more diversification. There are some overlaps with XSOE because of China, Taiwan and South Korea though.
Playing around with the country exposure in Excel gives me the following, this does not include the bond portion of NTSX.

While there are tax inefficiencies with US-based funds because of the dividend withholding taxes, I believe in timing the market using Good-After-Time Orders to avoid them.

Colony Capital (CLNY)
CLNY is a diversified REIT that holds/held a portfolio of Industrial, Hospitality, Healthcare and Commercial mortgage-backed securities (CMBS) which is transitioning towards a digital portfolio.
Source: Colony Capital Q2 2020 Earnings Presentation
Their strategy involves acquiring digital assets off the private markets and optimizing them to bring them to a lower multiple compared to their peers (DLR, EQIX, AMT, CCI, etc). Senior management has been overhauled which shows that they are serious about this change as well and not just doing mouth service. The new CEO, Marc Ganzi, has a history of working in digital and communicators assets also has a performance award of $100 m worth of shares if the stock price closes about $10 for 90 consecutive daysClark Value Street did a pretty detailed write up on the stock if you are interested.

The hospitality portfolio has recently been sold off, while older shareholders may not like it since they probably sold it a low(er) price given the current situation, new shareholders who bought in at current prices will probably appreciate it. Following that, the proceeds were use to redeem their 2021 convertible notes.
I would prefer if they redeemed the preferred shares as these are paying out >7% at par. There has been big volume on them following the sale of the hospitality portfolio but it could just be retailers and institutions expecting them to redeem the shares.

GameAccount Network Limited (GAN)
GAN provides software for online gambling and sports betting companies. Instead of buying individual online gambling companies, GAN gives you indirect exposure to them. In the last quarter, GAN has >100% revenue growth YoY and with large total accessible addressable market that is slowly opening up to them.
The share price took a hit after the earnings call on 20 August, and has been declining ever since. I took a position too early by writing puts around $18 and have a breakeven of about $16 after assignment but I am comfortable holding it.
Wanted to just go back to that for a second so just so we understand. The previous guidance of $37 million to $39 million, did that include FanDuel OSB revenue? And now that is being removed after August, so that's being replaced by, I guess, stronger non-FanDuel revenue, the recent launch of Cordish in PA and then Penn ramping in the back half. Is that the best way to think about it?
Yes. I mean, I think we've always been conservative relative to the sports forecast. A portion of that, of course, was the possibility again for FanDuel migrating off since they have the right under the contractual terms to do that. We knew that there would be other customers that we would be launching in the second half of the year to replace that as far as the 2020 revenue stream. So it was a balanced approach relative to the Q3, Q4 guidance, which is why we're not changing it at this point.
Source: Pennsylvania Gaming Control board
These 2 are GAN's clients which have recorded impressive growth. Some of it can be attributed to the 'lockdowns' and avoiding casinos. The risks involves their clients migrating to their own software liek FanDuel did and the online gambling 'boom' could just be a phase because of COVID19. Despite being online casinos, some of these companies have pretty low operating margins.

The BETZ ETF might be worth considering but they hold casinos with physical properties such as MGM, WYNN and PENN as well, though they also have their own online gaming offerings.

LLNW is an edge computing provider (buzword alert) with huge customer base that focuses on content delivery, live events and real time streaming. With cable cutting becoming more widespread, more households are turning towards streaming services such as Netflix (NFLX), Discovery (DISCA), HBO (T), Disney+ and Hulu (also DIS). The whole appeal of online streaming was convenience but they are now segregated again. LLNW is one of the beneficiaries since it benefits from the competition among these companies by being the backend provider. They also benefit from the increasing demand of online education.
They also completed a placement $110 m of 2025 convertible notes priced at $8.53 per share on 23 July 2020 and actively hiring
Source: Motley Fool Transcribing - Limelight Networks (LLNW) Q2 2020 Earnings Call Transcript
Amazon is a large client of theirs and their recent entry into live sports and events also benefits them, along with the launch of Amazon Luna. While others compare them with NET, FSLY or AKAM, LLNW has much smaller market cap the same as tailwinds as them. Most companies also depend on multiple companies to provide CDN and live streaming services as well. On a more speculative note, Bezos can easily buy them out at their current market cap.

KNSL and PLMR are both excess and surplus insurance providers which provide insurance coverage for things that are not covered by normal insurance providers. KNSL is has market cap of about 4 billion and PLMR at 2.5 billion. Both companies recorded impressive growth despite COVID19.
In addition to our own business strategy, our growth is being enhanced by a growing level of dislocation within the P&C market. After many years of intense competition, some competitors are experiencing adverse results and are withdrawing capacity, canceling some programs, raising prices, etc. We expect this dislocation to continue, thereby allowing Kinsale to grow at an elevated rate perhaps through 2021. At some point thereafter, we expect the level of dislocation to abate and our growth rate to normalize perhaps in the low double-digit range. Beyond the accelerated growth, industry dislocation is also allowing Kinsale to raise rates and in some cases, restrict coverage to further expand our profit margins. To take full advantage of this market opportunity, there is a possibility Kinsale could raise a modest amount of equity capital before year-end. At the end of the first quarter 2020, we noted that we did not expect the COVID-19 virus to have a material impact on Kinsale's profitability or growth. three months later, we have the exact same position.
PLMR has an even better combined ratio than KNSL and looks like the earlier days of KNSL in terms of growth.
Both companies trade at a very high P/E ratio and rightfully so because of their strong growth. 
Kinsale Q2 2020 Earnings Call Transcript
The risks of the businesses or losing clients to COVID19 is very real but KNSL seems pretty confident of continuing to grow in the current environment. The low interest rate environment is also bad for insurance companies as most of their investment portfolios contain of bonds. While Insurtech companies like Lemonade (LMND) or Shift Technologies may disrupt the market, the E&S market will probably be much harder for a smaller and new player to enter.
PG&E is the the sole provider of electricity in its coverage area. They recently emerged from their Chapter 11 with most of their assets and equity in place. Their dividends were suspended from 20 December 2017, after the California wildfires. Since this is an intended to be recovery play it doesn't really matter that much. Moreover, non-US residents such as myself are subject to the 30% dividend withholding tax so not having a dividend much be good for non-us investors.
As part of the Chapter 11 agreement, ~5.4 billion was paid out, funded partially by shares. The Fire Victim Trust now holds 24.6% of  the shareswhich as the fund for claimants to claim from. In a way, there is an incentive for them to want the share price increase so that more claims can be paid out. The lock up period for this should be over by now and its questionable what will happen to the share price. I don't think they'll just dump the shares and Institutions might buy them out of the fund's custodian. It'll definitely be hard for the company to improve EPS because of the diluted share base and the fact that utilities have large depreciation expense. Operating cash flow is all good though.
I have been writing $9 and $8.50 weekly-fortnightly puts from the $9-$9.50 range and it recently broke out. Its quite unfortunately but I already made ~$600 from my premiums collected which is pretty good since its about 10% gain with my normal position sizing.


Virtu Financial (VIRT)
VIRT is High Frequency Trading (HFT) and Market Making (MM) firm. With increased trading volumes and volatility, VIRT would be the equivalent of selling shovels during a gold rush.
Soruce: Virtu Financial September 2020 Presentation
Other alternatives are the brokers like IBKR or exchanges such as CBOE which looks rather attractive at the current price. The increase in trading volumes is most likely due to increased volatility and trader participation (Robinhooders + WFH) so I don't expect continued growth from VIRT in the long-term. It'll most likely end up as a dividend stock in the long run.

PRA Group (PRAA) & Encore Capital Group (ECPG)
Source: Encore Capital Group, Inc. Investor Presentation September 2020
These companies acquire non-performing loans from companies at steep discounts at proceed to recover them from the clients. Both companies have had a great Q2 results and with more bad debt, write offs and bankruptcies looming, this makes for an opportunistic play.

Liquidity Services (LQDT)
Source: Liquidity Services Investor Presentation Third Quarter Fiscal Year 2020
LQDT is reverse supply chain company, with more bankruptcies and cost savings, companies and liquidators will look to sell or acquire assets cheaply. Please read Irving's post on his blog for his analysis.

Friday, 4 September 2020

Frasers Centrepoint Trust (J69U) Acquisition of Shares in AsiaRetail Fund Limited and Proposed Divestment of Bedok Point

Frasers Centrepoint Trust already owns ~36.89% of the Asiamall Retail Fund (ARF). With this acquisition, they will own the entire 100% and it will be converted to an LLP which allows for REIT tax benefits, rather than being taxed as it is currently held as investment. This transaction also includes a divestment of a Malaysia mall that is held in the ARF.

Both properties are being divested to the Sponsor (Frasers Property (TQ5)).

The funds will be raised from a mix of private placement and preferential offers. The illustrative price of the new units are priced at $2.22, making the adjusted price about $2.42 at the based on the last close.
Price based on 3rd September 2020
Outstanding units from ShareInvestor
The private placement price might be $2.22 so retail investors may end up getting a higher subscription price.
With the completion of the acquisition, the portfolio is much more 'heartland'. I'm still not a fan of Changi City Point though.
The Pro forma DPU increases due to the divestment of Bedok Point. Without Bedok Point, its actually not accretive in the current environment. It's a very good acquisition without the rental rebates but this will depend on how long government support and COVID will last. Will landlords still be able to command the same or even higher rentals going forward?
Bedok Point being divested to the sponsor may create some form of conflict of interest. I don't really understand the terms but it most likely will allow the sponsor to inject the property back to FCT at a higher valuation if the property value increases, or at a lower price (but higher valuation) if the property valuations decreases.

Anyway, the acquisition of the ARF was expected for some time as Frasers Property kept increasing their stake it in and acquiring the manager. This might also be a good time for them to divest while property valuations are still high. While the heartland malls are packed, the requirement of social distancing eats into the profit margins of F&B businesses which may require landlords to lower rents to help them. It also feels like the impact of recession is not being felt yet.