Last month, we discussed the millennial problem in the context of the declining economics of gambling (See The Millennial Problem: Why We (Don't) Gamble). As we noted, both Nevada and New Jersey have in the past year enacted legislation which will allow new types of skill-based games, which many in the gaming industry are hoping will improve the draw of the (dealerless) gaming floor to such millennials.

Late September's Global Gaming Expo (G2E) at Las Vegas Sands' Sands Expo and Convention Center in Las Vegas offered a look at the first wave of such skill-based games. These games ranged from low-risk, no-brainers – such as IGT's (NYSE: IGT) Texas Tea Pinball and Scientifc Games' (NASDAQ: SGMS) Space Invaders – to more radical plays like NanoTech Gaming's Vegas Pinball 2047, as well as other ideas which are likely in need of work.

Let's start by talking about the best skill-based game in the show.

Best in Show: Grab Poker (Gamblit Gaming)

During a keynote rountable discussion Wednesday, new Caesars Entertainment (NASDAQ: CZR) CEO Mark Frissora noted that O'Sheas was "crazy good," and with its 10 beer pong tables is the highest grossing bar at the LINQ Hotel & Casino on the Las Vegas Strip. While it's unclear how big of a hurdle the other bars at the LINQ present, you get the point: O'Sheas is an attractive place for millennials to hang out and perhaps gamble a bit on $5 tables, and is one of the few places on the Strip where the cover bands regularly play the '90s music that older millennials (such as myself) grew up on.

Gamblit Gaming's Grab Poker is a game that embodies the spirit of the discussion about millennials, and represents the type of game innovation that will appeal to millennials and fit this type of environment.

Grab Poker is a multi-player poker game played on an electronic table top. To begin, 2-4 players sit or stand around the table, and buy in for $1 each (or some other amount). Each player starts with two initial cards, with the ultimate goal of completing the best five-card poker hand at the table before time runs out. Cards then appear in the center of the table one-by-one in rapid succession – if a card appears that helps your hand, you want to "grab" the card by pressing a button (or rather a designated area) before someone else takes it, or before the next card appears.

When the first player completes a five-card hand, a countdown is initiated, and the remaining players try to improve their hands with whatever cards come before time runs out. The player with the best hand wins the pot.

Grab Poker is simple, and the kind of game you can grab a beer and sit down and play. It's loads of fun, and versatile enough to stick just about anywhere, from the bar to the casino floor. You can play with your friends, or make new ones.

At G2E, the demo was set to take a 15% rake. The game takes 20% from each player off the top; when four players buy in for $1 each, the final pot is $3.20 rather than $4. But the game also returns another 5% in the form of gold tokens that randomly appear in the game in the center of the table and are up for grabs – these tokens had $5, $20, and $100 denominations for the demo, though I was told that other variable denominations may be used.

The cost per play on a $1 buy-in winds up being about $0.15 per person on average, assuming all players are equally skilled.

The first question is whether or not people would play this game for money. My initial thought was "I don't know." But then my next thought was, I would definitely pay to play the game for fun, much the same way I would pay to play air hockey in an arcade. And if I would pay to play the game in an arcade, then of course I would play for $1 with a chance to get some of my money back, or even win.

And that to me is the right context to think about this game – not so much as a gambling game, but rather as an arcade game in which you might get some of your money back or even win.

The game does have some potential limitations, chiefly the possibility of collusion. For example, let's say we have a player (Player A) who is playing with a buddy (Player B) and a third player (Player C). Player A has trips and the best hand, while his buddy (Player B) is drawing dead (can't win) and the third player is drawing at a diamond flush; now instead of playing to win, Player B might try to grab any diamonds to block the third player from making a flush, and thus improve his buddy's chance of winning.

The possibility of collusion is probably not a big deal if we're talking about $1 games played among friendly people, but might be a bigger issue if we're talking about higher-stakes games, and may preclude such games from occurring. And I might be cautious about walking up to a table where two guys are just sitting around not playing, and waiting for somebody else to show up.

I'd also be concerned that a casino operator might take the wrong approach and either rake too much or make the stakes too high to be played casually. At $1 buy-in, the player will spend $0.15 on average for a game that lasts about 45 seconds; this is a relatively nominal wager and expenditure, and the gambling aspect is mostly de-emphasized.

However, the way the demo was set up, the game is generating $0.60 in revenue in those 45 seconds assuming a full four players – less if only two or three people play (though the games are also shorter with fewer players). If a casino operator decides that's not enough, it might be tempted to raise either the rake and/or the buy-in. But 15% is already pushing it as far as rake rate, while buy-in sizes that start at $2 per game might start to look too much like a serious gambling game where the result matters – rather than a casual arcade game – which might limit the market for the game, and for which a 15% rake is way too high.

That said, if the goal of the casino operator is to broaden its market and appeal more to millennials, this is the type of game that may help do the trick.

No-Brainers: Texas Tea Pinball (IGT) and Space Invaders (Scientific Games)

IGT's Texas Tea Pinball and Scientific Games' Space Invaders are both standard video slot games with skill-based bonus rounds. These games are low-risk, no-brainers for the casino operator because the player can simply opt out of the skill-based challenges and take free spins instead.

This allows the games to appeal to players interested in the skill-based bonus games without excluding traditional video slot players.

IGT's Texas Tea Pinball is the flashier of the two. The game starts as a standard video slot; when the bonus is triggered, a screen pops up where the player can choose either a traditional free game bonus or to play the pinball bonus in which the player's score in a pinball game affects the final payout. This screen also displays the potential reward for advancing levels in the pinball game; the player's goal is to score enough points to advance levels in order to achieve the highest payout.

When the player completes the pinball bonus, a wheel appears; the player spins the wheel for a final payout. This mechanism appears to serve to neutralize the impact of skill, such that a player who screws up the pinball game does not get overpenalized.

I'd play it.

Similarly, in Scientific Games' Space Invaders, a player triggering the bonus has the option of choosing either free games or trying his skill at playing the arcade classic Space Invaders. In the bonus game, the player is rewarded for racking up points shooting bad guys before his cannon is destroyed; in addition, the player also achieves a higher payout for advancing levels, as in IGT's Texas Tea Pinball. To be honest, I didn't have much luck (or skill) at this, but I suspect the bonus feature will find an audience, while the Space Invaders theme, throwback graphics, and the ability to opt out will have appeal for non-skillers as well.

Radical Plays: Vegas Pinball 2047 (NanoTech Gaming)

Las Vegas-based NanoTech Gaming made a bit of a splash last year when it introduced its Vegas Pinball 2047 game at G2E 2014. Vegas Pinball 2047 bypasses slot play entirely and is a straight pinball game played on a nifty "steampunk"-styled pinball cabinet with physical buttons for gameplay, but on an electronic screen.

The pinball game itself requires no explanation; the key is how wagering is implemented, because it is quite complex, while the outcome is still largely random.

The first thing you have to do is set the size of your wager and desired prize. You can, for example, bet $100 to win $10 and win with a high frequency, or perhaps bet $5 to win $200 and win your bet with a low frequency.

The way the wagering is resolved is that at the end of the game, a color wheel appears with green and red slices. If you select a high win frequency wager (such as when you bet $100 to win $10) the green slices will dominate the wheel; if you select a low win frequency wager (such as betting $5 to win $200) the wheel will be dominated by red slices.

You then spin the wheel. If the pointer lands on green, you win; if it lands on red, you lose. Meanwhile, your skill at pinball impacts the outcome in that the better your performance at pinball, the bigger the green area is on the wheel, and thus the better your chances of winning.

This year, along with Vegas Pinball 2047, NanoTech also brought a Pac-Man clone called CasinoKat, which uses the same wagering mechanism.

When Vegas Pinball 2047 was introduced at G2E 2014, it was presented as an "Advantage Play High Limit Pinball" game, designed to be beatable by highly skilled pinballers and with $100 minimum wagers.

I don't see either of those things happening. For starters, I don't see casino operators offering a game that's beatable – particularly one where a single advantage player can hog the machine and prevent non-advantage players from playing. Nor do I see casual players playing this for $100 a pop.

I do think these games are different, and that there's probably room for something like these games, if not these games specifically as constructed. The underlying games themselves are no-brainers, as everybody already knows how to play pinball and Pac-Man. However, the implementation of wagering – having the player set his bets, how the wheel at the end of the game randomizes the final win/loss outcome, and the extent to which game play affects the randomizing wheel at the end of the game – will require a bit of coaching and will likely present a hurdle.

I don't fancy the chances of the average person sitting or standing at one of these machines and figuring out how the wagering works on his own.

In fact, the wagering is even more complex than what I've described because these games also utilize a technology that allows for a floating house advantage. The simplest way to explain this is that the casino can set the target payback percentage to 99% (1% house advantage); in order for an advantage player to be able to play at 101% payback (1% player advantage), somebody else then has to play poorly such that they play at 97% payback. If a lot of bad players play, then the achievable payback rate rises; if a lot of good players play, then the achievable payback rate falls so that the casino gets its 99% overall payback rate.

So in actuality, a good player can't just hog the machine and prevent bad players from playing; instead, they can hog the machine when the achievable payback rate is high, and then leave the worse conditions for players who don't know the difference. Now I'm not a legal expert, but even if a casino operator wanted to offer such a proposition, I'm skeptical that regulations allowing such a mechanism would gain widespread adoption as more states look at skill-based gaming.

My recommendation would be to ditch the mechanism that allows for a floating house advantage, and also to ditch the variable betting option and standardize on a fixed bet size and prize. This could be something like $5 minimum bets with a 1-to-1 payoff (a $5 win for a $5 bet) and a 10% house advantage, which equates to a minimum player spend of $0.50 per play – somewhere in the neighborhood of what one might spend on a pinball machine in an arcade. Such a proposition would be more palatable, and a reasonable fit for a casual, millennial-friendly place like O'Sheas.

Needs Work: Most Everything Else

Many of the other games presented at G2E feature clones of games which have proven to be popular as non-gambling games, but where the implementation of wagering may be flawed, due to either:

1. Logistical issues, and/or

2. A disconnect between the game and the wagering.

In a standard slot game, a wager typically occurs on every spin, except for bonus rounds. But in at least one game presented at G2E, the player alternates between one round where the player makes a wager on a slot game, and then one much longer round where the player plays the skill game with no wagering or monetary prizes involved. And in many of the other "skill-based" games presented at G2E, the wagering takes place not on the game itself, but rather when the player completes an action, such as killing a bad guy in Doom (the first-person shooter and PC classic), destroying a structure with a tank, or matching fruit or animal symbols.

The first problem is logistical. If the player is only wagering intermittently, then what happens is that you are creating dead space where no wagering (i.e. spending) occurs. Consequently, the casino operator is going to want to compensate for the lost wagering either by raising the stakes or the house edge – if not both – whenever wagering occurs. And as we established in our discussion of the millennial problem, we are already at a point where gambling value stinks, and increasingly so.

Giving the gambler an even worse proposition than before is not a solution.

The other problem is that there is often a disconnect between the wagering and the game itself. In many of the games presented, the wagering component winds up being completely arbitrary, while also penalizing the player for achieving what should be positive results.

When I think about playing Doom for money, for example, I envision something like this: I make a $5 buy-in; if I kill enough bad guys and/or achieve certain goals such as finishing a level in a certain time frame, I expect to be rewarded, and perhaps win more than I bought in for.

But that's not at all what's happening, as you don't make a wager on the game itself. Instead:

1. You set a bet amount, such as $0.50 or $1.00.

2. You pick up the controller and play the game; no wagering has occurred yet.

3. You kill a bad guy – triggering a wager based on your pre-determined bet size – and a point total shows up indicating the result of your wager on that particular kill. That point total may be more or less than the amount of your wager.

The outcome of the game itself at least appears to be mostly irrelevant – the game is largely independent of the wagering, and vice versa.

In fact, it's worse than that. If I'm an intelligent person, I understand that wagering is a penalty – not a reward. Any wager I make will be with a negative expectation, and thus is an expenditure. If I'm playing Doom, I do not want to be "rewarded" for killing bad guys by being forced to make a minus-expected-value wager every time I kill one.

This wagering structure is not unique to the Doom game, but is common to many (if not the majority) of the "skill-based" games on display at G2E. And the likely motivations for structuring the wagering this way are: (1) to neutralize the impact of skill, where you don't want bad (unskilled) players getting crushed and thus avoiding the games entirely; and (2) to make the skill and wagering components distinct in order to potentially make it easier for such a game to pass regulation.

I should qualify the above by saying that there's nothing wrong with the underlying games. I have no problem sitting down and playing Doom, Guitar Hero, Candy Crush, or Battleship clones all day – many of us do that anyway.

But there's a lot of work that needs to be done implementing the wagering aspects of these games, with the challenge being doing so in the context of:

1. What will ultimately be legal jurisdiction by jurisdiction

2. Where an unskilled player who's never played the game in question isn't getting crushed, and

3. Where a skilled player doesn't have an advantage over the casino.

Final Thoughts

In all, I thought it was a pretty decent showing on the skill-based games front. At a minimum, we saw one major innovation (Gamblit's Grab Poker) I'm ready to go play now, as well as two more subtler innovations (IGT's Texas Tea Pinball and Scientific Games' Space Invaders) which should have little difficulty finding audiences.

What we haven't seen yet are completely new single-player skill games, in the way that video poker games are single-player skill games. Instead, most of what was shown are clones of popular non-gambling games with wagering technology slapped on, and not necessarily in a way that makes sense.

As a casual observer, it's important to realize that with regard to next-gen skill-based games, we are only talking about two states right now – Nevada and New Jersey. Moreover, despite all of the discussion regarding millennials, it's not even necessarily clear that we really even have a millennial "problem", while millennials are also still years away from being prime casino customers anyhow. Consequently, it shouldn't be surprising that the most aggressive innovations are coming from smaller upstarts like Gamblit Gaming, NanoTech Gaming, and G2 Game Design, while the larger companies (IGT and Scientfic Games) brought low-risk, no-brainer games that aren't a huge leap from the products they already offer.

The general consensus seems to be that we will begin to see skill-based games on the floor in late 2016 or early 2017, though one supplier I spoke to intimated that it had agreements in place for as early as Q2 2016, presumably for a field trial.