Two-Sided Markets Theory: Pricing Strategies and Network Effects

Game theory is the science of strategy that analyzes how people make decisions when outcomes depend not just on their own choices but also on what others choose to do. It examines the hidden rules behind real-world interactions between individuals, businesses, and nations. The basic elements include: players (decision makers), strategies (full set of choices), payoffs (outcomes), rules (what's allowed and known), and information (what each player knows). Game theory officially took shape in 1944 when John von Neumann and Oscar Morgenstern published 'The Theory of Games and Economic Behavior,' introducing zero-sum games where one person's gain is exactly another person's loss. In the 1950s, John Nash expanded it with the Nash equilibrium, a point where no player can improve their outcome by changing their strategy as long as others stay the same. The Prisoner's Dilemma captures a frustrating truth: sometimes doing what's best for yourself ends up being worse for everyone involved. Two people are arrested and separated, each offered the same deal: if one confesses while the other stays silent, the confessor goes free and the silent one gets a long sentence; if both confess, both get moderate sentences; if both stay silent, both get light sentences. Each person thinks, 'If the other stays quiet, I should confess and go free.' But if they confess, I'd better confess too, or I'll get the worst deal. This line of thinking leads both to confess even though staying silent would have given them both a better result.

International relations can be modeled using a simple game theory framework involving threats and responses. One actor decides whether to issue a threat; if no threat is issued, the status quo remains. If a threat is issued, the opponent chooses between escalation or concession. If the opponent escalates, the original actor must decide whether to follow through on punishment or back down. Three possible outcomes exist: opponent concession (good for initial actor, bad for opponent), opponent escalation without follow-through (worse than status quo for both due to economic uncertainty), and full escalation (worst outcome for both). When all actors are rational, backward induction shows that neither side should issue threats because the rational actor will ultimately back down, making the threat itself a losing proposition.

Hackenbush is a combinatorial game played on graphs with blue, red, and green branches connected to a ground node. Players take turns cutting branches of their own color or green branches, with disconnected components disappearing after each cut. The goal is to make the last move. The game was invented by John Conway in 1970. Positions are classified into four types: positive (Left wins regardless of who moves), negative (Right wins regardless), zero (second player wins), and fuzzy (confused with zero). Basic positions include +1 (Left move), -1 (Right move), 0 (no moves), and * (single green branch, losing for first player).

The fundamental recursive code structure for solving game theory problems (tasks 19-21) involves: (1) Reading the problem by diagonal - identifying two players (Petya, Vanya), one pile of stones (S), allowed moves (remove 2 stones or divide by 2), and termination condition (S ≤ 87); (2) Implementing a recursive function with parameters S (stones) and M (moves remaining); (3) Base cases: if S ≤ 87, check parity (odd = Petya wins, even = Vanya wins); if M = 0, return 0 (no winner); (4) Recursive step: return any from h if M is odd, else all from h; (5) Testing by iterating S from 89 to 200 and checking conditions like F(S,2) for Vanya winning first move.

This comprehensive tutorial covers the entire methodology for solving game theory problems. Students learn to systematically check for saddle points by comparing row maxima and column minima—when they match, pure strategies suffice; when not, mixed strategies are required. The video introduces probability concepts for mixed strategies, where choosing one strategy with probability x implies the other with probability (1-x). Key formulas for 2x2 games are presented: x = (a22 - a12)/(sum - cross), y = (a22 - a21)/(sum - cross), and game value = (a11×a22 - a12×a21)/(sum - cross). The dominance rule technique is taught for reducing larger matrices to 2x2 form: eliminate rows where all elements are greater than or equal to another row's elements, and eliminate columns where all elements are less than or equal to another column's elements. When pure dominance fails, students learn to calculate averages of strategy pairs and compare with remaining rows/columns. The complete workflow involves reducing matrices using dominance rules, rechecking for saddle points, and applying formulas to obtain optimal strategies and game values.

Traditional pricing involves affixed prices on items or shelves that remain constant for all customers. Dynamic pricing (also called elastic pricing) allows prices to change based on who is looking at the item, creating different prices for different shoppers at the same time. This model is legal in most states but raises concerns about fairness and transparency.

Traditional valuation models for cryptocurrencies, including those used by BlackRock in 2018, fundamentally fail to predict actual market prices. These models, which relied on network metrics like usage and volume, significantly underestimated XRP (predicting 30 cents or less) and Bitcoin (predicting $1,000 or less). The speaker demonstrates that no valuation model, including Cloud Code, could have predicted XRP's dramatic rallies in 2017 or 2024. Similarly, no analyst could have predicted Amazon's massive growth from 2006 based on linear analysis. This reveals that crypto and tech asset prices are driven by transformative events and narrative shifts rather than predictable patterns. XRP's unique value proposition lies in serving as a bridge currency for converting value between different points in the financial system, particularly in illiquid markets where direct conversion is impossible. Unlike stablecoins, XRP allows users to convert value with less capital, as price appreciation can achieve the same conversion value with reduced capital requirements.

The gaming industry's traditional pricing model has been fundamentally disrupted. For three decades, consumers could purchase consoles at premium launch prices, then buy facelifted versions cheaper after 2-3 years, or hardware-improved models at original prices. This implicit agreement has collapsed as PlayStation 5 prices have risen from $399-499 to $599-649, with the Pro model reaching $900. This shift means AAA games are becoming accessible primarily to wealthier consumers, potentially ending the traditional path to gaming for younger generations.

Traditional businesses that maintain historical pricing models often fail to account for inflation and changing cost structures. The Ratskeller's owner referenced that in the past, a meal cost 50 pfennigs, but today the same meal costs 5 euros. This represents a 100-fold increase in price, yet the restaurant continues to charge 5 euros for a complete lunch. The owner acknowledged that costs have changed dramatically, including the price of basic ingredients like cucumbers (now 80-90 cents each). This demonstrates that traditional pricing models must be regularly updated to reflect current economic realities.

Traditional retail pricing has become fundamentally unsustainable due to excessive intermediary costs. The price gap between physical stores and direct factory channels has reached 400% for electronics. Traditional retail prices include hidden costs for importers, distributors, wholesalers, and retailers—costs that don't add functional value. Harvard Business Research shows supply chain margins still bear 70% of final prices. China's response was the Manufacturer-to-Consumer (M2C) model, where factories like those in Shenzhen sell directly to consumers, eliminating the traditional 'arisan tengkulak' middleman network that inflated prices.

Network externalities, or network effects, are a type of positive externality where the value of a good or service increases as more people use it; unlike traditional goods where value decreases with more users (like sharing an apple), network goods like social media platforms (Facebook, WeChat) and ride-sharing services (Uber) become more valuable and useful as their user base grows, creating a self-reinforcing cycle where increased participation leads to greater individual utility.

Network externalities is the phenomenon where the value of a product or service is not determined by its physical characteristics or cost, but rather by how many people have access to it. The telephone example illustrates this: if you were the only person with a telephone, it would be useless and have no value. However, as soon as one other person has a telephone, your device gains value because you can now communicate with that person. The more people who have telephones, the more valuable your telephone becomes because you can call more people. This concept applies to any product or service where its value increases with the number of users.

Network externalities (also called network effects) is a situation where the value and utility of a good or service to a user depends on how many other people use it. The usefulness of a product depends on the size of its user network.

Positive network externalities occur when increased adoption by others raises individual demand. Fashion exemplifies this: as more people adopt a style, it becomes more desirable for others to follow, creating self-reinforcing popularity cycles. Telecommunications similarly demonstrate this principle, as phones become more valuable with larger user bases enabling broader communication networks. This phenomenon is also called the bandwagon effect, where consumers follow popular trends to gain social acceptance and utility.

Network externalities occur when the demand for a good depends on how many other people use it. Positive network externalities (bandwagon effect) increase demand as more people use the good. Negative network externalities (snob effect) decrease demand as more people use the good. This creates interdependence among consumers' purchasing decisions.