Pickers
Part hidden-object, part negotiation simulator: this American Pickers game will have you searching for valuables and selling them for the best price. Buy low and sell high, baby! Players engage in negotiations to buy items for the lowest possible price, which is encompassed in a hidden-object a meta-game. The player and the AI will go back and forth offering prices until the player accepts the AI’s offer, the AI accepts the player’s, or the deal falls flat entirely.
One of my goals with this project was to semi-realistically model the mechanics of negotiation. I’ll try to explain the AI’s reactions here. The item’s base value will be randomly assigned. The seller’s starting price is a random percentage higher than the item’s base value, and the seller’s minimum price is a random percentage lower than the item’s base value. The player can either press the “Deal” button (which accepts the current price), or type a number into the counter-offer input field and press the “Make Offer” button to negotiate. If the offer is greater than the asking price, then the seller instantly accepts the deal. If it is less than a threshold below the minimum price, then the seller does not change his price. If the offer is above the minimum threshold and below a threshold under the asking price, the seller will lower its current asking price. If the offer is very near to the current asking price, the AI will accept the offer.
Another mechanic I explored was price obfuscation. By default, I do not reveal the item’s value to the player. They can only see the rough range of the value (Example: "Item Value: Hmm... That's probably worth around a thousand"). By obfuscating the item’s exact value, the player is forced to guess the item’s value based on the value range and the seller’s asking price; and because the player always make the first offer, I am able to invoke in the player a strategy/feeling of “hmm… maybe I should lowball him.”
During playtesting, I found that most people tended to avoid the “Big ticket items” (Items worth more than $5,000) and searched for the lower-tier items. This was a problem because I found that the negotiations were more engaging when dealing with larger denominations of money. A major cause of this, of course, was that during playtesting there was no way to actually regain spent money – which has since been implemented. To avoid the “Big Ticket” stigma I chose to not actually change any item value, but to make the value ranges smaller. Instead of “Big Ticket” items being those worth over $5000, they are now defined as items worth over $7,500. This cuts the amount of “Bit Ticket” items in half without actually modifying the game’s balance.
If I were to continue work on this project, I would greatly expand its scope. Ideally, it would have multiple locations and an antique shop metagame. Instead of instantly receiving money for the item’s value, players would instead need to sell them in their store at a marked up price. Such a metagame would give players a greater since of purpose rather than “buy all the things!”
One of my goals with this project was to semi-realistically model the mechanics of negotiation. I’ll try to explain the AI’s reactions here. The item’s base value will be randomly assigned. The seller’s starting price is a random percentage higher than the item’s base value, and the seller’s minimum price is a random percentage lower than the item’s base value. The player can either press the “Deal” button (which accepts the current price), or type a number into the counter-offer input field and press the “Make Offer” button to negotiate. If the offer is greater than the asking price, then the seller instantly accepts the deal. If it is less than a threshold below the minimum price, then the seller does not change his price. If the offer is above the minimum threshold and below a threshold under the asking price, the seller will lower its current asking price. If the offer is very near to the current asking price, the AI will accept the offer.
Another mechanic I explored was price obfuscation. By default, I do not reveal the item’s value to the player. They can only see the rough range of the value (Example: "Item Value: Hmm... That's probably worth around a thousand"). By obfuscating the item’s exact value, the player is forced to guess the item’s value based on the value range and the seller’s asking price; and because the player always make the first offer, I am able to invoke in the player a strategy/feeling of “hmm… maybe I should lowball him.”
During playtesting, I found that most people tended to avoid the “Big ticket items” (Items worth more than $5,000) and searched for the lower-tier items. This was a problem because I found that the negotiations were more engaging when dealing with larger denominations of money. A major cause of this, of course, was that during playtesting there was no way to actually regain spent money – which has since been implemented. To avoid the “Big Ticket” stigma I chose to not actually change any item value, but to make the value ranges smaller. Instead of “Big Ticket” items being those worth over $5000, they are now defined as items worth over $7,500. This cuts the amount of “Bit Ticket” items in half without actually modifying the game’s balance.
If I were to continue work on this project, I would greatly expand its scope. Ideally, it would have multiple locations and an antique shop metagame. Instead of instantly receiving money for the item’s value, players would instead need to sell them in their store at a marked up price. Such a metagame would give players a greater since of purpose rather than “buy all the things!”
|
|